---
title: "Locate, Steer, and Improve — paper by Zhang et al. (2026)"
description: "Most earlier reviews of mechanistic interpretability (MI) in LLMs describe what researchers have… 392 references, cited by 9 papers here."
authors: ["Hengyuan Zhang", "Zhihao Zhang", "Ercong Nie", "Mingyang Wang", "Zunhai Su", "Yiwei Wang", "Qianli Wang", "Shuzhou Yuan", "Xufeng Duan", "Qibo Xue", "Zeping Yu", "Chenming Shang", "Xiao Liang", "Jing Xiong", "Hui Shen", "Chaofan Tao", "Zhengwu Liu", "Senjie Jin", "Zhiheng Xi", "Dongdong Zhang", "Sophia Ananiadou", "Tao Gui", "Ruobing Xie", "Hayden Kwok-Hay So", "Hinrich Schuetze", "Xuanjing Huang", "Qi Zhang", "Ngai Wong"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.502/"
doi: "10.18653/v1/2026.findings-acl.502"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/zhang-2026-locate-steer-improve-practical-survey/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/zhang-2026-locate-steer-improve-practical-survey/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/2/) / Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models

# Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models

**Authors:** [Hengyuan Zhang](https://md.research.elamur.ai/free/en/authors/hengyuan-zhang/), [Zhihao Zhang](https://md.research.elamur.ai/free/en/authors/zhihao-zhang/), [Ercong Nie](https://md.research.elamur.ai/free/en/authors/ercong-nie/), [Mingyang Wang](https://md.research.elamur.ai/free/en/authors/mingyang-wang/), [Zunhai Su](https://md.research.elamur.ai/free/en/authors/zunhai-su/), [Yiwei Wang](https://md.research.elamur.ai/free/en/authors/yiwei-wang/), [Qianli Wang](https://md.research.elamur.ai/free/en/authors/qianli-wang/), [Shuzhou Yuan](https://md.research.elamur.ai/free/en/authors/shuzhou-yuan/), [Xufeng Duan](https://md.research.elamur.ai/free/en/authors/xufeng-duan/), Qibo Xue, [Zeping Yu](https://md.research.elamur.ai/free/en/authors/zeping-yu/), [Chenming Shang](https://md.research.elamur.ai/free/en/authors/chenming-shang/), [Xiao Liang](https://md.research.elamur.ai/free/en/authors/xiao-liang/), [Jing Xiong](https://md.research.elamur.ai/free/en/authors/jing-xiong/), Hui Shen, [Chaofan Tao](https://md.research.elamur.ai/free/en/authors/chaofan-tao/), Zhengwu Liu, [Senjie Jin](https://md.research.elamur.ai/free/en/authors/senjie-jin/), [Zhiheng Xi](https://md.research.elamur.ai/free/en/authors/zhiheng-xi/), [Dongdong Zhang](https://md.research.elamur.ai/free/en/authors/dongdong-zhang/), [Sophia Ananiadou](https://md.research.elamur.ai/free/en/authors/sophia-ananiadou/), [Tao Gui](https://md.research.elamur.ai/free/en/authors/tao-gui/), [Ruobing Xie](https://md.research.elamur.ai/free/en/authors/ruobing-xie/), [Hayden Kwok-Hay So](https://md.research.elamur.ai/free/en/authors/hayden-kwok-hay-so/), [Hinrich Schuetze](https://md.research.elamur.ai/free/en/authors/hinrich-schuetze/), [Xuanjing Huang](https://md.research.elamur.ai/free/en/authors/xuanjing-huang/), [Qi Zhang](https://md.research.elamur.ai/free/en/authors/qi-zhang/) and [Ngai Wong](https://md.research.elamur.ai/free/en/authors/ngai-wong/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.502/) · [PDF](https://aclanthology.org/2026.findings-acl.502.pdf) · [DOI 10.18653/v1/2026.findings-acl.502](https://doi.org/10.18653/v1/2026.findings-acl.502) · [OpenAlex](https://openalex.org/W7166846670)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [9](https://md.research.elamur.ai/free/en/papers/zhang-2026-locate-steer-improve-practical-survey/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Most earlier reviews of mechanistic interpretability (MI) in LLMs describe what researchers have discovered inside models. They give less help to someone who wants to act on those findings. This survey reframes MI as an engineering toolkit organized as a three-stage pipeline, "Locate, Steer, and Improve." It first gives formal definitions of the interpretable components of decoder-only Transformers, such as the residual stream, attention heads, FFN neurons, SAE features and circuits. It then groups diagnostic (localizing) and intervention (steering) techniques according to which of these components they act on. For each technique it states the formulation, the components it applies to, and practical trade-offs such as compute cost and data dependence. The authors then map the literature onto three improvement goals: alignment, capability and efficiency. For each scenario they summarize typical usage patterns. The central claim is that many recent advances, including safety alignment, knowledge editing and sparse fine-tuning, work best when they rest on explicit mechanistic interventions. This matters because model growth driven only by scaling is reaching its limits.

## Key points
- **Localizing methods (six families):**
  - Magnitude Analysis: needs no training but depends on input data.
  - Causal Attribution: patching and ablation.
  - Gradient Detection.
  - Probing.
  - Vocabulary Projection: Logit Lens.
  - Circuit Discovery.
- **Steering methods (three families):**
  - Amplitude Manipulation: scaling or replacing activations.
  - Targeted Optimization: weight editing and targeted fine-tuning.
  - Vector Arithmetic: feature and task vectors.
- **Application areas:**
  - Alignment: safety and reliability, fairness and bias, persona and role.
  - Capability: reasoning, multilingualism, knowledge management.
  - Efficiency: efficient training and efficient inference.
- **Efficiency examples:**
  - Fine-tuning only very sparse subsets, such as about 0.1% of neurons found via causal attribution.
  - Pruning tokens and KV cache.
  - Skipping redundant layers or MoE experts.
  - Layer-wise mixed-precision quantization guided by how sensitive each layer is.
- **Resources and limitations:**
  - The authors provide a method-tagged list of more than 200 papers in the GitHub repo Awesome-Actionable-MI-Survey.
  - Coverage is limited to dense LLMs; MoE and vision models are not systematically treated.
  - The authors note that there are no unified benchmarks for evaluating localization methods.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Microsoft](https://md.research.elamur.ai/free/en/companies/microsoft/) — Company, Redmond, United States · 378 works · 161,630 citations
- [Tsinghua University](https://md.research.elamur.ai/free/en/universities/tsinghua-university/) — University, Beijing, China · 415 works · 94,028 citations
- [Technische Universität Darmstadt](https://md.research.elamur.ai/free/en/universities/technische-universitat-darmstadt/) — University, Darmstadt, Germany · 114 works · 36,326 citations
- [Fudan University](https://md.research.elamur.ai/free/en/universities/fudan-university/) — University, Shanghai, China · 202 works · 22,509 citations
- [Tencent](https://md.research.elamur.ai/free/en/companies/tencent/) — Company, Shenzhen, China · 163 works · 21,032 citations
- [Ludwig-Maximilians-Universität München](https://md.research.elamur.ai/free/en/universities/ludwig-maximilians-universitat-munchen/) — University, Munich, Germany · 113 works · 56,817 citations
- [Chinese University of Hong Kong](https://md.research.elamur.ai/free/en/universities/chinese-university-of-hong-kong/) — University, Hong Kong, China · 159 works · 16,154 citations
- [University of California, Los Angeles](https://md.research.elamur.ai/free/en/universities/university-of-california-los-angeles/) — University, Los Angeles, United States · 99 works · 56,333 citations
- [University of Hong Kong](https://md.research.elamur.ai/free/en/universities/university-of-hong-kong/) — University, Pok Fu Lam, China · 118 works · 13,496 citations
- [University of Manchester](https://md.research.elamur.ai/free/en/universities/university-of-manchester/) — University, Manchester, United Kingdom · 93 works · 45,138 citations
- [Nanjing University](https://md.research.elamur.ai/free/en/universities/nanjing-university/) — University, Nanjing, China · 99 works · 8,512 citations
- [Dartmouth College](https://md.research.elamur.ai/free/en/universities/dartmouth-college/) — University, Hanover, United States · 41 works · 42,719 citations
- [Technische Universität Dresden](https://md.research.elamur.ai/free/en/universities/technische-universitat-dresden/) — University, Dresden, Germany · 13 works · 7,336 citations
- [Technische Universität Berlin](https://md.research.elamur.ai/free/en/universities/technische-universitat-berlin/) — University, Berlin, Germany · 22 works · 957 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works · [🔬 Interpretability](https://md.research.elamur.ai/free/en/topics/interpretability/) · 149 works

## Cite

Cite as: Hengyuan Zhang, Zhihao Zhang, Ercong Nie, Mingyang Wang, Zunhai Su, Yiwei Wang, Qianli Wang, Shuzhou Yuan, Xufeng Duan, Qibo Xue, Zeping Yu, Chenming Shang, Xiao Liang, Jing Xiong, Hui Shen, Chaofan Tao, Zhengwu Liu, Senjie Jin, Zhiheng Xi, Dongdong Zhang, Sophia Ananiadou, Tao Gui, Ruobing Xie, Hayden Kwok-Hay So, Hinrich Schuetze, Xuanjing Huang, Qi Zhang and Ngai Wong. 2026. Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{zhang2026locate,
  title = {{Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models}},
  author = {Hengyuan Zhang, Zhihao Zhang, Ercong Nie, Mingyang Wang, Zunhai Su, Yiwei Wang, Qianli Wang, Shuzhou Yuan, Xufeng Duan, Qibo Xue, Zeping Yu, Chenming Shang, Xiao Liang, Jing Xiong, Hui Shen, Chaofan Tao, Zhengwu Liu, Senjie Jin, Zhiheng Xi, Dongdong Zhang, Sophia Ananiadou, Tao Gui, Ruobing Xie, Hayden Kwok-Hay So, Hinrich Schuetze, Xuanjing Huang, Qi Zhang and Ngai Wong},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.502},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.502}
}
```

- [← previous LearnAct: Few-Shot Mobile GUI Agent with a Unified Demonstration Benchmark](https://md.research.elamur.ai/free/en/papers/liu-2026-learnact-few-shot-mobile-gui/)
- [next → MAWARITH: A Dataset and Benchmark for Legal Inheritance Reasoning with LLMs](https://md.research.elamur.ai/free/en/papers/bouchekif-2026-mawarith-dataset-benchmark-legal-inheritance/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/2/)

## References

392 references, 80 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Marah Abdin, Jyoti Aneja, Hany Awadalla, Ahmed Awadallah, Ammar Ahmad Awan, Nguyen Bach, Amit Bahree, Arash Bakhtiari, Jianmin Bao, Harkirat Behl, Alon Benhaim, Misha Bilenko, Johan Bjorck, Sébastien Bubeck, Martin Cai, Qin Cai, Vishrav Chaudhary, Dong Chen, Dongdong Chen, and 110 others. 2024. Phi-… — [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://md.research.elamur.ai/free/en/papers/abdin-2024-phi-3-technical-report-highly/)
2. Hiba Ahsan, Arnab Sen Sharma, Silvio Amir, David Bau, and Byron C. Wallace. 2025. Elucidating Mechanisms of Demographic Bias in LLMs for Healthcare. http://arxiv.org/abs/2502.13319. arXiv:2502.13319. [↗](http://arxiv.org/abs/2502.13319)
3. Preprint, Mst. Shapna Akter, Hossain Shahriar, Alfredo Cuzzocrea, and Fan Wu. 2024. Uncovering the interpretation of large language models. In 48th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2024, Osaka, Japan, July 2-4, 2024, pages 1057–1066. IEEE. [find](https://scholar.google.com/scholar?q=Preprint%2C+Mst%2E+Shapna+Akter%2C+Hossain+Shahriar%2C+Alfredo+Cuzzocrea%2C+and+Fan+Wu%2E+2024%2E+Uncovering+the+interpretation+of+large+language+models%2E+In+48th+IEEE+Annual+Computers%2C+Software%2C+and+Applications+Co)
4. Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, Roman Ring, Eliza Rutherford, Serkan Cabi, Tengda Han, Zhitao Gong, Sina Samangooei, Marianne Monteiro, Jacob L. Menick, Sebastian Borgeaud, and 8… — [Flamingo: a Visual Language Model for Few-Shot Learning](https://md.research.elamur.ai/free/en/papers/alayrac-2022-flamingo-visual-language-model-few/)
5. Zeyuan Allen-Zhu and Yuanzhi Li. 2023. Physics of language models: Part 1, learning hierarchical language structures. arXiv preprint arXiv:2305.13673. [find](https://scholar.google.com/scholar?q=Zeyuan+Allen%2DZhu+and+Yuanzhi+Li%2E+2023%2E+Physics+of+language+models%3A+Part+1%2C+learning+hierarchical+language+structures%2E+arXiv+preprint+arXiv%3A2305%2E13673%2E)
6. Emmanuel Ameisen, Jack Lindsey, Adam Pearce, Wes Gurnee, Nicholas L. Turner, Brian Chen, Craig Citro, David Abrahams, Shan Carter, Basil Hosmer, Jonathan Marcus, Michael Sklar, Adly Templeton, Trenton Bricken, Callum McDougall, Hoagy Cunningham, Thomas Henighan, Adam Jermyn, Andy Jones, and 8 others… [find](https://scholar.google.com/scholar?q=Emmanuel+Ameisen%2C+Jack+Lindsey%2C+Adam+Pearce%2C+Wes+Gurnee%2C+Nicholas+L%2E+Turner%2C+Brian+Chen%2C+Craig+Citro%2C+David+Abrahams%2C+Shan+Carter%2C+Basil+Hosmer%2C+Jonathan+Marcus%2C+Michael+Sklar%2C+Adly+Templeton%2C+Trenton)
7. Yongqi An, Xu Zhao, Tao Yu, Ming Tang, and Jinqiao Wang. 2025. Systematic outliers in large language models. arXiv preprint arXiv:2502.06415. — [arXiv:2502.06415](https://arxiv.org/abs/2502.06415) [find](https://scholar.google.com/scholar?q=Yongqi+An%2C+Xu+Zhao%2C+Tao+Yu%2C+Ming+Tang%2C+and+Jinqiao+Wang%2E+2025%2E+Systematic+outliers+in+large+language+models%2E+arXiv+preprint+arXiv%3A2502%2E06415%2E+%E2%80%94+%5BarXiv%3A2502%2E06415%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2502%2E06415%29)
8. Lyzander Marciano Andrylie, Inaya Rahmanisa, Mahardika Krisna Ihsani, Alfan Farizki Wicaksono, Haryo Akbarianto Wibowo, and Alham Fikri Aji. 2025. Sparse autoencoders can capture language-specific concepts across diverse languages. Preprint, arXiv:2507.11230. — [arXiv:2507.11230](https://arxiv.org/abs/2507.11230) [find](https://scholar.google.com/scholar?q=Lyzander+Marciano+Andrylie%2C+Inaya+Rahmanisa%2C+Mahardika+Krisna+Ihsani%2C+Alfan+Farizki+Wicaksono%2C+Haryo+Akbarianto+Wibowo%2C+and+Alham+Fikri+Aji%2E+2025%2E+Sparse+autoencoders+can+capture+language%2Dspecific+con)
9. Anthropic. 2024. Introducing Claude 3.5 Sonnet. Announcement of Claude 3.5 Sonnet model release, featuring improved intelligence, vision capabilities, and new Artifacts feature. [find](https://scholar.google.com/scholar?q=Anthropic%2E+2024%2E+Introducing+Claude+3%2E5+Sonnet%2E+Announcement+of+Claude+3%2E5+Sonnet+model+release%2C+featuring+improved+intelligence%2C+vision+capabilities%2C+and+new+Artifacts+feature%2E)
10. Dana Arad, Aaron Mueller, and Yonatan Belinkov. 2025. [Saes are good for steering–if you select the right features](https://md.research.elamur.ai/free/en/papers/arad-2025-saes-are-good-steering-if/). arXiv preprint arXiv:2505.20063.
11. Andy Arditi, Oscar Balcells Obeso, Aaquib Syed, Daniel Paleka, Nina Rimsky, Wes Gurnee, and Neel Nanda. 2024. [Refusal in language models is mediated by a single direction](https://md.research.elamur.ai/free/en/papers/arditi-2024-refusal-language-models-mediated-single/). In The Thirty-eighth Annual Conference on Neural Information Processing Systems.
12. Saleh Ashkboos, Amirkeivan Mohtashami, Maximilian L Croci, Bo Li, Pashmina Cameron, Martin Jaggi, Dan Alistarh, Torsten Hoefler, and James Hensman. 2024. [Quarot: Outlier-free 4-bit inference in rotated llms](https://md.research.elamur.ai/free/en/papers/ashkboos-2024-quarot-outlier-free-4-bit/). Advances in Neural Information Processing Systems, 37:100213–100240.
13. Behrooz Azarkhalili and Maxwell W. Libbrecht. 2025. Generalized attention flow: Feature attribution for transformer models via maximum flow. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2025, Vienna, Austria, July 27 - August… [find](https://scholar.google.com/scholar?q=Behrooz+Azarkhalili+and+Maxwell+W%2E+Libbrecht%2E+2025%2E+Generalized+attention+flow%3A+Feature+attribution+for+transformer+models+via+maximum+flow%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the+Associatio)
14. Nooshin Bahador. 2025. Mechanistic interpretability of fine-tuned vision transformers on distorted images: Decoding attention head behavior for transparent and trustworthy ai. arXiv preprint arXiv:2503.18762. — [arXiv:2503.18762](https://arxiv.org/abs/2503.18762) [find](https://scholar.google.com/scholar?q=Nooshin+Bahador%2E+2025%2E+Mechanistic+interpretability+of+fine%2Dtuned+vision+transformers+on+distorted+images%3A+Decoding+attention+head+behavior+for+transparent+and+trustworthy+ai%2E+arXiv+preprint+arXiv%3A250)
15. Sarah Ball, Frauke Kreuter, and Nina Panickssery. 2024. [Understanding jailbreak success: A study of latent space dynamics in large language models](https://md.research.elamur.ai/free/en/papers/ball-2026-understanding-jailbreak-success-study-latent/). Preprint, arXiv:2406.09289.
16. Amin Banayeeanzade, Ala N Tak, Fatemeh Bahrani, Anahita Bolourani, Leonardo Blas, Emilio Ferrara, Jonathan Gratch, and Sai Praneeth Karimireddy. 2025. Psychological steering in llms: An evaluation of effectiveness and trustworthiness. arXiv preprint arXiv:2510.04484. [find](https://scholar.google.com/scholar?q=Amin+Banayeeanzade%2C+Ala+N+Tak%2C+Fatemeh+Bahrani%2C+Anahita+Bolourani%2C+Leonardo+Blas%2C+Emilio+Ferrara%2C+Jonathan+Gratch%2C+and+Sai+Praneeth+Karimireddy%2E+2025%2E+Psychological+steering+in+llms%3A+An+evaluation+of+)
17. Tetiana Bas and Krystian Novak. 2025. Steering latent traits, not learned facts: An empirical study of activation control limits. arXiv preprint arXiv:2511.18284. — [arXiv:2511.18284](https://arxiv.org/abs/2511.18284) [find](https://scholar.google.com/scholar?q=Tetiana+Bas+and+Krystian+Novak%2E+2025%2E+Steering+latent+traits%2C+not+learned+facts%3A+An+empirical+study+of+activation+control+limits%2E+arXiv+preprint+arXiv%3A2511%2E18284%2E+%E2%80%94+%5BarXiv%3A2511%2E18284%5D%28https%3A%2F%2Farxiv%2Eor)
18. Reza Bayat, Ali Rahimi-Kalahroudi, Mohammad Pezeshki, Sarath Chandar, and Pascal Vincent. 2025. Steering large language model activations in sparse spaces. arXiv preprint arXiv:2503.00177. — [arXiv:2503.00177](https://arxiv.org/abs/2503.00177) [find](https://scholar.google.com/scholar?q=Reza+Bayat%2C+Ali+Rahimi%2DKalahroudi%2C+Mohammad+Pezeshki%2C+Sarath+Chandar%2C+and+Pascal+Vincent%2E+2025%2E+Steering+large+language+model+activations+in+sparse+spaces%2E+arXiv+preprint+arXiv%3A2503%2E00177%2E+%E2%80%94+%5BarXiv%3A25)
19. Eden Biran, Daniela Gottesman, Sohee Yang, Mor Geva, and Amir Globerson. 2024. [Hopping too late: Exploring the limitations of large language models on multi-hop queries](https://md.research.elamur.ai/free/en/papers/biran-2024-hopping-too-late-exploring-limitations/). In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA, November 12…
20. Joseph Bloom. 2024. Open source sparse autoencoders for all residual stream layers of gpt2 small. [find](https://scholar.google.com/scholar?q=Joseph+Bloom%2E+2024%2E+Open+source+sparse+autoencoders+for+all+residual+stream+layers+of+gpt2+small%2E)
21. Yelysei Bondarenko, Markus Nagel, and Tijmen Blankevoort. 2023. Quantizable transformers: Removing outliers by helping attention heads do nothing. In Advances in Neural Information Processing Systems, volume 36, pages 75067–75096. Curran Associates, Inc. [find](https://scholar.google.com/scholar?q=Yelysei+Bondarenko%2C+Markus+Nagel%2C+and+Tijmen+Blankevoort%2E+2023%2E+Quantizable+transformers%3A+Removing+outliers+by+helping+attention+heads+do+nothing%2E+In+Advances+in+Neural+Information+Processing+Systems%2C)
22. Joschka Braun, Carsten Eickhoff, and Seyed Ali Bahrainian. 2025. Beyond multiple choice: Evaluating steering vectors for adaptive free-form summarization. arXiv preprint arXiv:2505.24859. — [arXiv:2505.24859](https://arxiv.org/abs/2505.24859) [find](https://scholar.google.com/scholar?q=Joschka+Braun%2C+Carsten+Eickhoff%2C+and+Seyed+Ali+Bahrainian%2E+2025%2E+Beyond+multiple+choice%3A+Evaluating+steering+vectors+for+adaptive+free%2Dform+summarization%2E+arXiv+preprint+arXiv%3A2505%2E24859%2E+%E2%80%94+%5BarXiv%3A250)
23. Trenton Bricken and Cengiz Pehlevan. 2021. Attention approximates sparse distributed memory. Advances in Neural Information Processing Systems, 34:15301– 15315. [find](https://scholar.google.com/scholar?q=Trenton+Bricken+and+Cengiz+Pehlevan%2E+2021%2E+Attention+approximates+sparse+distributed+memory%2E+Advances+in+Neural+Information+Processing+Systems%2C+34%3A15301%E2%80%93+15315%2E)
24. Trenton Bricken, Adly Templeton, Joshua Batson, Brian Chen, Adam Jermyn, Tom Conerly, Nick Turner, Cem Anil, Carson Denison, Amanda Askell, Robert Lasenby, Yifan Wu, Shauna Kravec, Nicholas Schiefer, Tim Maxwell, Nicholas Joseph, Zac Hatfield-Dodds, Alex Tamkin, Karina Nguyen, and 6 others. 2023. To… [find](https://scholar.google.com/scholar?q=Trenton+Bricken%2C+Adly+Templeton%2C+Joshua+Batson%2C+Brian+Chen%2C+Adam+Jermyn%2C+Tom+Conerly%2C+Nick+Turner%2C+Cem+Anil%2C+Carson+Denison%2C+Amanda+Askell%2C+Robert+Lasenby%2C+Yifan+Wu%2C+Shauna+Kravec%2C+Nicholas+Schiefer%2C+)
25. Nora Belrose, Zach Furman, Logan Smith, Danny Halawi, Igor Ostrovsky, Lev McKinney, Stella Biderman, and Jacob Steinhardt. 2023. [Eliciting latent predictions from transformers with the tuned lens](https://md.research.elamur.ai/free/en/papers/belrose-2023-eliciting-latent-predictions-transformers-tuned/). arXiv preprint arXiv:2303.08112.
26. Jannik Brinkmann, Chris Wendler, Christian Bartelt, and Aaron Mueller. 2025. Large language models share representations of latent grammatical concepts across typologically diverse languages. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computat… [find](https://scholar.google.com/scholar?q=Jannik+Brinkmann%2C+Chris+Wendler%2C+Christian+Bartelt%2C+and+Aaron+Mueller%2E+2025%2E+Large+language+models+share+representations+of+latent+grammatical+concepts+across+typologically+diverse+languages%2E+In+Proce)
27. Leonard Bereska and Efstratios Gavves. 2024. [Mechanistic interpretability for ai safety–a review](https://md.research.elamur.ai/free/en/papers/bereska-2024-mechanistic-interpretability-ai-safety-review/). arXiv preprint arXiv:2404.14082.
28. Bart Bussmann, Patrick Leask, and Neel Nanda. 2024. Batchtopk sparse autoencoders. arXiv preprint arXiv:2412.06410. — [arXiv:2412.06410](https://arxiv.org/abs/2412.06410) [find](https://scholar.google.com/scholar?q=Bart+Bussmann%2C+Patrick+Leask%2C+and+Neel+Nanda%2E+2024%2E+Batchtopk+sparse+autoencoders%2E+arXiv+preprint+arXiv%3A2412%2E06410%2E+%E2%80%94+%5BarXiv%3A2412%2E06410%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2412%2E06410%29)
29. Sumanta Bhattacharyya and Pedram Rooshenas. 2025. Steered generation via gradient descent on sparse features. arXiv preprint arXiv:2502.18644. — [arXiv:2502.18644](https://arxiv.org/abs/2502.18644) [find](https://scholar.google.com/scholar?q=Sumanta+Bhattacharyya+and+Pedram+Rooshenas%2E+2025%2E+Steered+generation+via+gradient+descent+on+sparse+features%2E+arXiv+preprint+arXiv%3A2502%2E18644%2E+%E2%80%94+%5BarXiv%3A2502%2E18644%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2502%2E18644%29)
30. Yuchen Cai, Ding Cao, Rongxi Guo, Yaqin Wen, Guiquan Liu, and Enhong Chen. 2024a. Locating and Mitigating Gender Bias in Large Language Models. Preprint, arXiv:2403.14409. — [arXiv:2403.14409](https://arxiv.org/abs/2403.14409) [find](https://scholar.google.com/scholar?q=Yuchen+Cai%2C+Ding+Cao%2C+Rongxi+Guo%2C+Yaqin+Wen%2C+Guiquan+Liu%2C+and+Enhong+Chen%2E+2024a%2E+Locating+and+Mitigating+Gender+Bias+in+Large+Language+Models%2E+Preprint%2C+arXiv%3A2403%2E14409%2E+%E2%80%94+%5BarXiv%3A2403%2E14409%5D%28https%3A%2F)
31. Jing Bi, Junjia Guo, Yunlong Tang, Lianggong Bruce Wen, Zhang Liu, Bingjie Wang, and Chenliang Xu. 2025. Unveiling visual perception in language models: An attention head analysis approach. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 4135–4144. [find](https://scholar.google.com/scholar?q=Jing+Bi%2C+Junjia+Guo%2C+Yunlong+Tang%2C+Lianggong+Bruce+Wen%2C+Zhang+Liu%2C+Bingjie+Wang%2C+and+Chenliang+Xu%2E+2025%2E+Unveiling+visual+perception+in+language+models%3A+An+attention+head+analysis+approach%2E+In+Proceed)
32. Zefan Cai, Yichi Zhang, Bofei Gao, Yuliang Liu, Yucheng Li, Tianyu Liu, Keming Lu, Wayne Xiong, Yue Dong, Junjie Hu, and 1 others. 2024b. [Pyramidkv: Dynamic kv cache compression based on pyramidal information funneling](https://md.research.elamur.ai/free/en/papers/cai-2024-pyramidkv-dynamic-kv-cache-compression/). arXiv preprint arXiv:2406.02069.
33. Yonatan Belinkov. 2022. [Probing classifiers: Promises, shortcomings, and advances](https://md.research.elamur.ai/free/en/papers/belinkov-2022-probing-classifiers-promises-shortcomings-advances/). Comput. Linguistics, 48(1):207–219.
34. Nicola Cancedda. 2024. Spectral filters, dark signals, and attention sinks. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4792– 4808. [find](https://scholar.google.com/scholar?q=Nicola+Cancedda%2E+2024%2E+Spectral+filters%2C+dark+signals%2C+and+attention+sinks%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28Volume+1%3A+Long+Papers%29%2C+pages+47)
35. Bhavik Chandna, Zubair Bashir, and Procheta Sen. 2025. Dissecting Bias in LLMs: A Mechanistic Interpretability Perspective. Preprint, arXiv:2506.05166. — [arXiv:2506.05166](https://arxiv.org/abs/2506.05166) [find](https://scholar.google.com/scholar?q=Bhavik+Chandna%2C+Zubair+Bashir%2C+and+Procheta+Sen%2E+2025%2E+Dissecting+Bias+in+LLMs%3A+A+Mechanistic+Interpretability+Perspective%2E+Preprint%2C+arXiv%3A2506%2E05166%2E+%E2%80%94+%5BarXiv%3A2506%2E05166%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2506%2E)
36. Yuan Chang, Ziyue Li, Hengyuan Zhang, Yuanbo Kong, Yanru Wu, Hayden Kwok-Hay So, Zhijiang Guo, Liya Zhu, and Ngai Wong. 2025. [Treereview: A dynamic tree of questions framework for deep and efficient llm-based scientific peer review](https://md.research.elamur.ai/free/en/papers/chang-2025-treereview-dynamic-tree-questions-framework/). In Proceedings of the 2025 Conference on Empirical Methods in Natur…
37. David Chanin, James Wilken-Smith, Tomás Dulka, Hardik Bhatnagar, and Joseph Bloom. 2024. A is for absorption: Studying feature splitting and absorption in sparse autoencoders. CoRR, abs/2409.14507. — [arXiv:2409.14507](https://arxiv.org/abs/2409.14507) [find](https://scholar.google.com/scholar?q=David+Chanin%2C+James+Wilken%2DSmith%2C+Tom%C3%A1s+Dulka%2C+Hardik+Bhatnagar%2C+and+Joseph+Bloom%2E+2024%2E+A+is+for+absorption%3A+Studying+feature+splitting+and+absorption+in+sparse+autoencoders%2E+CoRR%2C+abs%2F2409%2E14507%2E+%E2%80%94+)
38. Alan Chen, Jack Merullo, Alessandro Stolfo, and Ellie Pavlick. 2025a. Transferring linear features across language models with model stitching. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Alan+Chen%2C+Jack+Merullo%2C+Alessandro+Stolfo%2C+and+Ellie+Pavlick%2E+2025a%2E+Transferring+linear+features+across+language+models+with+model+stitching%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Informat)
39. Jianhui Chen, Xiaozhi Wang, Zijun Yao, Yushi Bai, Lei Hou, and Juanzi Li. 2025b. Towards understanding safety alignment: A mechanistic perspective from safety neurons. [find](https://scholar.google.com/scholar?q=Jianhui+Chen%2C+Xiaozhi+Wang%2C+Zijun+Yao%2C+Yushi+Bai%2C+Lei+Hou%2C+and+Juanzi+Li%2E+2025b%2E+Towards+understanding+safety+alignment%3A+A+mechanistic+perspective+from+safety+neurons%2E)
40. Lihu Chen, Adam Dejl, and Francesca Toni. 2025c. Identifying query-relevant neurons in large language models for long-form texts. In AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA, pages 23595–23604. AAAI Press… [find](https://scholar.google.com/scholar?q=Lihu+Chen%2C+Adam+Dejl%2C+and+Francesca+Toni%2E+2025c%2E+Identifying+query%2Drelevant+neurons+in+large+language+models+for+long%2Dform+texts%2E+In+AAAI%2D25%2C+Sponsored+by+the+Association+for+the+Advancement+of+Artifi)
41. Runjin Chen, Andy Arditi, Henry Sleight, Owain Evans, and Jack Lindsey. 2025d. Persona vectors: Monitoring and controlling character traits in language models. arXiv preprint arXiv:2507.21509. — [arXiv:2507.21509](https://arxiv.org/abs/2507.21509) [find](https://scholar.google.com/scholar?q=Runjin+Chen%2C+Andy+Arditi%2C+Henry+Sleight%2C+Owain+Evans%2C+and+Jack+Lindsey%2E+2025d%2E+Persona+vectors%3A+Monitoring+and+controlling+character+traits+in+language+models%2E+arXiv+preprint+arXiv%3A2507%2E21509%2E+%E2%80%94+%5BarXi)
42. Shiqi Chen, Miao Xiong, Junteng Liu, Zhengxuan Wu, Teng Xiao, Siyang Gao, and Junxian He. 2024a. In-context sharpness as alerts: An inner representation perspective for hallucination mitigation. In Forty-first International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Shiqi+Chen%2C+Miao+Xiong%2C+Junteng+Liu%2C+Zhengxuan+Wu%2C+Teng+Xiao%2C+Siyang+Gao%2C+and+Junxian+He%2E+2024a%2E+In%2Dcontext+sharpness+as+alerts%3A+An+inner+representation+perspective+for+hallucination+mitigation%2E+In+Fo)
43. Wei Chen, Zhen Huang, Liang Xie, Binbin Lin, Houqiang Li, Le Lu, Xinmei Tian, Deng Cai, Yonggang Zhang, Wenxiao Wang, Xu Shen, and Jieping Ye. 2024b. From yes-men to truth-tellers: Addressing sycophancy in large language models with pinpoint tuning. In Forty-first International Conference on Machine… [find](https://scholar.google.com/scholar?q=Wei+Chen%2C+Zhen+Huang%2C+Liang+Xie%2C+Binbin+Lin%2C+Houqiang+Li%2C+Le+Lu%2C+Xinmei+Tian%2C+Deng+Cai%2C+Yonggang+Zhang%2C+Wenxiao+Wang%2C+Xu+Shen%2C+and+Jieping+Ye%2E+2024b%2E+From+yes%2Dmen+to+truth%2Dtellers%3A+Addressing+sycophan)
44. Xinrong Chen, Xu Chu, Yingmin Qiu, Hengyuan Zhang, Jing Xiong, Shiyu Tang, Shuai Liu, Shaokang Yang, Cheng Yang, Hayden Kwok-Hay So, and 1 others. 2026. Residual decoding: Mitigating hallucinations in large vision-language models via history-aware residual guidance. arXiv preprint arXiv:2602.01047. — [arXiv:2602.01047](https://arxiv.org/abs/2602.01047) [find](https://scholar.google.com/scholar?q=Xinrong+Chen%2C+Xu+Chu%2C+Yingmin+Qiu%2C+Hengyuan+Zhang%2C+Jing+Xiong%2C+Shiyu+Tang%2C+Shuai+Liu%2C+Shaokang+Yang%2C+Cheng+Yang%2C+Hayden+Kwok%2DHay+So%2C+and+1+others%2E+2026%2E+Residual+decoding%3A+Mitigating+hallucinations+in)
45. Yilong Chen, Junyuan Shang, Zhenyu Zhang, Yanxi Xie, Jiawei Sheng, Tingwen Liu, Shuohuan Wang, Yu Sun, Hua Wu, and Haifeng Wang. 2025e. [Inner thinking transformer: Leveraging dynamic depth scaling to foster adaptive internal thinking](https://md.research.elamur.ai/free/en/papers/chen-2025-inner-thinking-transformer-leveraging-dynamic/). In Proceedings of the 63rd Annual Meeting of the Association for…
46. Yuheng Chen, Pengfei Cao, Yubo Chen, Kang Liu, and Jun Zhao. 2024c. Journey to the center of the knowledge neurons: Discoveries of language-independent knowledge neurons and degenerate knowledge neurons. In Thirty-Eighth AAAI Conference on Artificial Intelligence, AAAI 2024, Thirty-Sixth Conference… [find](https://scholar.google.com/scholar?q=Yuheng+Chen%2C+Pengfei+Cao%2C+Yubo+Chen%2C+Kang+Liu%2C+and+Jun+Zhao%2E+2024c%2E+Journey+to+the+center+of+the+knowledge+neurons%3A+Discoveries+of+language%2Dindependent+knowledge+neurons+and+degenerate+knowledge+neuro)
47. Yuheng Chen, Pengfei Cao, Yubo Chen, Kang Liu, and Jun Zhao. 2025f. Knowledge localization: Mission not accomplished? enter query localization! In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Yuheng+Chen%2C+Pengfei+Cao%2C+Yubo+Chen%2C+Kang+Liu%2C+and+Jun+Zhao%2E+2025f%2E+Knowledge+localization%3A+Mission+not+accomplished%3F+enter+query+localization%21+In+The+Thirteenth+International+Conference+on+Learning+R)
48. Zewen Chi, Heyan Huang, and Xian-Ling Mao. 2023. Can cross-lingual transferability of multilingual transformers be activated without end-task data? In Findings of the Association for Computational Linguistics: ACL 2023, pages 12572–12584, Toronto, Canada. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Zewen+Chi%2C+Heyan+Huang%2C+and+Xian%2DLing+Mao%2E+2023%2E+Can+cross%2Dlingual+transferability+of+multilingual+transformers+be+activated+without+end%2Dtask+data%3F+In+Findings+of+the+Association+for+Computational+Lin)
49. Abhijith Chintam, Rahel Beloch, Willem Zuidema, Michael Hanna, and Oskar Van Der Wal. 2023. [Identifying and Adapting Transformer-Components Responsible for Gender Bias in an English Language Model](https://md.research.elamur.ai/free/en/papers/chintam-2023-identifying-adapting-transformer-components-responsible/). In Proceedings of the 6th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pa…
50. Hakaze Cho, Haolin Yang, Brian M Kurkoski, and Naoya Inoue. 2025. Binary autoencoder for mechanistic interpretability of large language models. arXiv preprint arXiv:2509.20997. — [arXiv:2509.20997](https://arxiv.org/abs/2509.20997) [find](https://scholar.google.com/scholar?q=Hakaze+Cho%2C+Haolin+Yang%2C+Brian+M+Kurkoski%2C+and+Naoya+Inoue%2E+2025%2E+Binary+autoencoder+for+mechanistic+interpretability+of+large+language+models%2E+arXiv+preprint+arXiv%3A2509%2E20997%2E+%E2%80%94+%5BarXiv%3A2509%2E20997%5D%28ht)
51. Ikhyun Cho and Julia Hockenmaier. 2025. Toward efficient sparse autoencoder-guided steering for improved in-context learning in large language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 28949–28961. [find](https://scholar.google.com/scholar?q=Ikhyun+Cho+and+Julia+Hockenmaier%2E+2025%2E+Toward+efficient+sparse+autoencoder%2Dguided+steering+for+improved+in%2Dcontext+learning+in+large+language+models%2E+In+Proceedings+of+the+2025+Conference+on+Empirica)
52. Yung-Sung Chuang, Yujia Xie, Hongyin Luo, Yoon Kim, James R. Glass, and Pengcheng He. 2024. [Dola: Decoding by contrasting layers improves factuality in large language models](https://md.research.elamur.ai/free/en/papers/chuang-2023-dola-decoding-contrasting-layers-improves/). In The Twelfth International Conference on Learning Representations.
53. Henry Conklin and Kenny Smith. 2024. Representations as language: An information-theoretic framework for interpretability. arXiv:2406.02449. arXiv preprint Arthur Conmy, Augustine Mavor-Parker, Aengus Lynch, Stefan Heimersheim, and Adrià Garriga-Alonso. 2023. Towards automated circuit discovery for… [find](https://scholar.google.com/scholar?q=Henry+Conklin+and+Kenny+Smith%2E+2024%2E+Representations+as+language%3A+An+information%2Dtheoretic+framework+for+interpretability%2E+arXiv%3A2406%2E02449%2E+arXiv+preprint+Arthur+Conmy%2C+Augustine+Mavor%2DParker%2C+Aengus)
54. Justin Cui, Wei-Lin Chiang, Ion Stoica, and Cho-Jui Hsieh. 2024. [OR-Bench: An over-refusal benchmark for large language models](https://md.research.elamur.ai/free/en/papers/cui-2024-bench-over-refusal-benchmark-large/). arXiv preprint arXiv:2405.20947.
55. Hoagy Cunningham, Aidan Ewart, Logan Riggs, Robert Huben, and Lee Sharkey. 2023. Sparse autoencoders find highly interpretable features in language models. arXiv preprint arXiv:2309.08600. [find](https://scholar.google.com/scholar?q=Hoagy+Cunningham%2C+Aidan+Ewart%2C+Logan+Riggs%2C+Robert+Huben%2C+and+Lee+Sharkey%2E+2023%2E+Sparse+autoencoders+find+highly+interpretable+features+in+language+models%2E+arXiv+preprint+arXiv%3A2309%2E08600%2E)
56. Bartosz Cywiński, Bart Bussmann, Arthur Conmy, Josh Engels, Neel Nanda, and Senthooran Rajamanoharan. 2025. Can we interpret latent reasoning using current mechanistic interpretability tools? [find](https://scholar.google.com/scholar?q=Bartosz+Cywin%CC%81ski%2C+Bart+Bussmann%2C+Arthur+Conmy%2C+Josh+Engels%2C+Neel+Nanda%2C+and+Senthooran+Rajamanoharan%2E+2025%2E+Can+we+interpret+latent+reasoning+using+current+mechanistic+interpretability+tools%3F)
57. Patrick Queiroz Da Silva, Hari Sethuraman, Dheeraj Rajagopal, Hannaneh Hajishirzi, and Sachin Kumar. 2025. Steering off course: Reliability challenges in steering language models. arXiv preprint arXiv:2504.04635. [find](https://scholar.google.com/scholar?q=Patrick+Queiroz+Da+Silva%2C+Hari+Sethuraman%2C+Dheeraj+Rajagopal%2C+Hannaneh+Hajishirzi%2C+and+Sachin+Kumar%2E+2025%2E+Steering+off+course%3A+Reliability+challenges+in+steering+language+models%2E+arXiv+preprint+arXiv)
58. Damai Dai, Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, and Furu Wei. 2022. Knowledge neurons in pretrained transformers. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2022, Dublin, Ireland, May 22-27, 2022, pages 8493– 8502.… [find](https://scholar.google.com/scholar?q=Damai+Dai%2C+Li+Dong%2C+Yaru+Hao%2C+Zhifang+Sui%2C+Baobao+Chang%2C+and+Furu+Wei%2E+2022%2E+Knowledge+neurons+in+pretrained+transformers%2E+In+Proceedings+of+the+60th+Annual+Meeting+of+the+Association+for+Computationa)
59. Adam Davies and Ashkan Khakzar. 2024. The cognitive revolution in interpretability: From explaining behavior to interpreting representations and algorithms. arXiv preprint arXiv:2408.05859. — [arXiv:2408.05859](https://arxiv.org/abs/2408.05859) [find](https://scholar.google.com/scholar?q=Adam+Davies+and+Ashkan+Khakzar%2E+2024%2E+The+cognitive+revolution+in+interpretability%3A+From+explaining+behavior+to+interpreting+representations+and+algorithms%2E+arXiv+preprint+arXiv%3A2408%2E05859%2E+%E2%80%94+%5BarXiv%3A2)
60. DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, and 181 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in ll…](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/)
61. Jia Deng, Tianyi Tang, Yanbin Yin, Wenhao yang, Xin Zhao, and Ji-Rong Wen. 2025. Neuron based personality trait induction in large language models. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Jia+Deng%2C+Tianyi+Tang%2C+Yanbin+Yin%2C+Wenhao+yang%2C+Xin+Zhao%2C+and+Ji%2DRong+Wen%2E+2025%2E+Neuron+based+personality+trait+induction+in+large+language+models%2E+In+The+Thirteenth+International+Conference+on+Learni)
62. Fabrizio Dimino, Krati Saxena, Bhaskarjit Sarmah, and Stefano Pasquali. 2025. Tracing Positional Bias in Financial Decision-Making: Mechanistic Insights from Qwen2.5. In Proceedings of the 6th ACM International Conference on AI in Finance, pages 96–104. [find](https://scholar.google.com/scholar?q=Fabrizio+Dimino%2C+Krati+Saxena%2C+Bhaskarjit+Sarmah%2C+and+Stefano+Pasquali%2E+2025%2E+Tracing+Positional+Bias+in+Financial+Decision%2DMaking%3A+Mechanistic+Insights+from+Qwen2%2E5%2E+In+Proceedings+of+the+6th+ACM+Int)
63. Maximilian Dreyer, Lorenz Hufe, Jim Berend, Thomas Wiegand, Sebastian Lapuschkin, and Wojciech Samek. 2025. From what to how: Attributing clip’s latent components reveals unexpected semantic reliance. arXiv preprint arXiv:2505.20229. — [arXiv:2505.20229](https://arxiv.org/abs/2505.20229) [find](https://scholar.google.com/scholar?q=Maximilian+Dreyer%2C+Lorenz+Hufe%2C+Jim+Berend%2C+Thomas+Wiegand%2C+Sebastian+Lapuschkin%2C+and+Wojciech+Samek%2E+2025%2E+From+what+to+how%3A+Attributing+clip%E2%80%99s+latent+components+reveals+unexpected+semantic+reliance%2E)
64. Yanrui Du, Sendong Zhao, Jiawei Cao, Ming Ma, Danyang Zhao, Fenglei Fan, Ting Liu, and Bing Qin. 2024. Towards secure tuning: Mitigating security risks arising from benign instruction fine-tuning. CoRR, abs/2410.04524. — [arXiv:2410.04524](https://arxiv.org/abs/2410.04524) [find](https://scholar.google.com/scholar?q=Yanrui+Du%2C+Sendong+Zhao%2C+Jiawei+Cao%2C+Ming+Ma%2C+Danyang+Zhao%2C+Fenglei+Fan%2C+Ting+Liu%2C+and+Bing+Qin%2E+2024%2E+Towards+secure+tuning%3A+Mitigating+security+risks+arising+from+benign+instruction+fine%2Dtuning%2E+CoR)
65. Xufeng Duan, Xinyu Zhou, Bei Xiao, and Zhenguang Cai. 2025. [Unveiling language competence neurons: A psycholinguistic approach to model interpretability](https://md.research.elamur.ai/free/en/papers/duan-2025-unveiling-language-competence-neurons-psycholinguistic/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 10148–10157, Abu Dhabi, UAE. Association for Computational…
66. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, and 1 others. 2024. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/). arXiv e-prints, pages arXiv–2407.
67. Razvan-Gabriel Dumitru, Vikas Yadav, Rishabh Maheshwary, Paul-Ioan Clotan, Sathwik Tejaswi Madhusudhan, and Mihai Surdeanu. 2024. Layer-wise quantization: A pragmatic and effective method for quantizing llms beyond integer bit-levels. arXiv preprint arXiv:2406.17415. — [arXiv:2406.17415](https://arxiv.org/abs/2406.17415) [find](https://scholar.google.com/scholar?q=Razvan%2DGabriel+Dumitru%2C+Vikas+Yadav%2C+Rishabh+Maheshwary%2C+Paul%2DIoan+Clotan%2C+Sathwik+Tejaswi+Madhusudhan%2C+and+Mihai+Surdeanu%2E+2024%2E+Layer%2Dwise+quantization%3A+A+pragmatic+and+effective+method+for+quantizi)
68. Nelson Elhage, Tristan Hume, Catherine Olsson, Nicholas Schiefer, Tom Henighan, Shauna Kravec, Zac Hatfield-Dodds, Robert Lasenby, Dawn Drain, Carol Chen, Roger B. Grosse, Sam McCandlish, Jared Kaplan, Dario Amodei, Martin Wattenberg, and Christopher Olah. 2022. Toy models of superposition. CoRR, ab… [find](https://scholar.google.com/scholar?q=Nelson+Elhage%2C+Tristan+Hume%2C+Catherine+Olsson%2C+Nicholas+Schiefer%2C+Tom+Henighan%2C+Shauna+Kravec%2C+Zac+Hatfield%2DDodds%2C+Robert+Lasenby%2C+Dawn+Drain%2C+Carol+Chen%2C+Roger+B%2E+Grosse%2C+Sam+McCandlish%2C+Jared+Kaplan)
69. Nelson Elhage, Neel Nanda, Catherine Olsson, Tom Henighan, Nicholas Joseph, Ben Mann, Amanda Askell, Yuntao Bai, Anna Chen, Tom Conerly, Nova DasSarma, Dawn Drain, Deep Ganguli, Zac Hatfield-Dodds, Danny Hernandez, Andy Jones, Jackson Kernion, Liane Lovitt, Kamal Ndousse, and 6 others. 2021. A mathe… [find](https://scholar.google.com/scholar?q=Nelson+Elhage%2C+Neel+Nanda%2C+Catherine+Olsson%2C+Tom+Henighan%2C+Nicholas+Joseph%2C+Ben+Mann%2C+Amanda+Askell%2C+Yuntao+Bai%2C+Anna+Chen%2C+Tom+Conerly%2C+Nova+DasSarma%2C+Dawn+Drain%2C+Deep+Ganguli%2C+Zac+Hatfield%2DDodds%2C+Da)
70. Mostafa Elhoushi, Akshat Shrivastava, Diana Liskovich, Basil Hosmer, Bram Wasti, Liangzhen Lai, Anas Mahmoud, Bilge Acun, Saurabh Agarwal, Ahmed Roman, Ahmed Aly, Beidi Chen, and Carole-Jean Wu. 2024. [LayerSkip: Enabling early exit inference and self-speculative decoding](https://md.research.elamur.ai/free/en/papers/elhoushi-2024-layerskip-enabling-early-exit-inference/). In Proceedings of the 62nd…
71. William Fedus, Barret Zoph, and Noam Shazeer. 2022. Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity. J. Mach. Learn. Res., 23:120:1–120:39. [find](https://scholar.google.com/scholar?q=William+Fedus%2C+Barret+Zoph%2C+and+Noam+Shazeer%2E+2022%2E+Switch+transformers%3A+Scaling+to+trillion+parameter+models+with+simple+and+efficient+sparsity%2E+J%2E+Mach%2E+Learn%2E+Res%2E%2C+23%3A120%3A1%E2%80%93120%3A39%2E)
72. Jiahai Feng and Jacob Steinhardt. 2023. How do language models bind entities in context? In NeurIPS 2023 Workshop on Symmetry and Geometry in Neural Representations. [find](https://scholar.google.com/scholar?q=Jiahai+Feng+and+Jacob+Steinhardt%2E+2023%2E+How+do+language+models+bind+entities+in+context%3F+In+NeurIPS+2023+Workshop+on+Symmetry+and+Geometry+in+Neural+Representations%2E)
73. Javier Ferrando, Gabriele Sarti, Arianna Bisazza, and Marta R Costa-Jussà. 2024. [A primer on the inner workings of transformer-based language models](https://md.research.elamur.ai/free/en/papers/ferrando-2024-primer-inner-workings-transformer-based/). arXiv preprint arXiv:2405.00208.
74. Javier Ferrando and Elena Voita. 2024. [Information flow routes: Automatically interpreting language models at scale](https://md.research.elamur.ai/free/en/papers/ferrando-2024-information-flow-routes-automatically-interpreting/). arXiv preprint arXiv:2403.00824.
75. Pedro Ferreira, Wilker Aziz, and Ivan Titov. 2025. Truthful or fabricated? using causal attribution to mitigate reward hacking in explanations. In Workshop on Actionable Interpretability at ICML 2025. [find](https://scholar.google.com/scholar?q=Pedro+Ferreira%2C+Wilker+Aziz%2C+and+Ivan+Titov%2E+2025%2E+Truthful+or+fabricated%3F+using+causal+attribution+to+mitigate+reward+hacking+in+explanations%2E+In+Workshop+on+Actionable+Interpretability+at+ICML+2025%2E)
76. Hiroki Furuta, Gouki Minegishi, Yusuke Iwasawa, and Yutaka Matsuo. 2024. Towards empirical interpretation of internal circuits and properties in grokked transformers on modular polynomials. Transactions on Machine Learning Research. [find](https://scholar.google.com/scholar?q=Hiroki+Furuta%2C+Gouki+Minegishi%2C+Yusuke+Iwasawa%2C+and+Yutaka+Matsuo%2E+2024%2E+Towards+empirical+interpretation+of+internal+circuits+and+properties+in+grokked+transformers+on+modular+polynomials%2E+Transactio)
77. Andrey Galichin, Alexey Dontsov, Polina Druzhinina, Anton Razzhigaev, Oleg Y Rogov, Elena Tutubalina, and Ivan Oseledets. 2025. I have covered all the bases here: Interpreting reasoning features in large language models via sparse autoencoders. arXiv preprint arXiv:2503.18878. — [arXiv:2503.18878](https://arxiv.org/abs/2503.18878) [find](https://scholar.google.com/scholar?q=Andrey+Galichin%2C+Alexey+Dontsov%2C+Polina+Druzhinina%2C+Anton+Razzhigaev%2C+Oleg+Y+Rogov%2C+Elena+Tutubalina%2C+and+Ivan+Oseledets%2E+2025%2E+I+have+covered+all+the+bases+here%3A+Interpreting+reasoning+features+in+la)
78. Sandeep Reddy Gantla. 2025. Exploring mechanistic interpretability in large language models: Challenges, approaches, and insights. In 2025 International Conference on Data Science, Agents & Artificial Intelligence (ICDSAAI), pages 1–8. IEEE. [find](https://scholar.google.com/scholar?q=Sandeep+Reddy+Gantla%2E+2025%2E+Exploring+mechanistic+interpretability+in+large+language+models%3A+Challenges%2C+approaches%2C+and+insights%2E+In+2025+International+Conference+on+Data+Science%2C+Agents+%26+Artificial)
79. Cheng Gao, Huimin Chen, Chaojun Xiao, Zhiyi Chen, Zhiyuan Liu, and Maosong Sun. 2025a. H-neurons: On the existence, impact, and origin of hallucination-associated neurons in llms. Preprint, arXiv:2512.01797. — [arXiv:2512.01797](https://arxiv.org/abs/2512.01797) [find](https://scholar.google.com/scholar?q=Cheng+Gao%2C+Huimin+Chen%2C+Chaojun+Xiao%2C+Zhiyi+Chen%2C+Zhiyuan+Liu%2C+and+Maosong+Sun%2E+2025a%2E+H%2Dneurons%3A+On+the+existence%2C+impact%2C+and+origin+of+hallucination%2Dassociated+neurons+in+llms%2E+Preprint%2C+arXiv%3A2512)
80. Leo Gao, Tom Dupré la Tour, Henk Tillman, Gabriel Goh, Rajan Troll, Alec Radford, Ilya Sutskever, Jan Leike, and Jeffrey Wu. 2024. Scaling and evaluating sparse autoencoders. arXiv preprint arXiv:2406.04093. [find](https://scholar.google.com/scholar?q=Leo+Gao%2C+Tom+Dupr%C3%A9+la+Tour%2C+Henk+Tillman%2C+Gabriel+Goh%2C+Rajan+Troll%2C+Alec+Radford%2C+Ilya+Sutskever%2C+Jan+Leike%2C+and+Jeffrey+Wu%2E+2024%2E+Scaling+and+evaluating+sparse+autoencoders%2E+arXiv+preprint+arXiv%3A2406)
81. Leo Gao, Achyuta Rajaram, Jacob Coxon, Soham V Govande, Bowen Baker, and Dan Mossing. 2025b. Weight-sparse transformers have interpretable circuits. arXiv preprint arXiv:2511.13653. — [arXiv:2511.13653](https://arxiv.org/abs/2511.13653) [find](https://scholar.google.com/scholar?q=Leo+Gao%2C+Achyuta+Rajaram%2C+Jacob+Coxon%2C+Soham+V+Govande%2C+Bowen+Baker%2C+and+Dan+Mossing%2E+2025b%2E+Weight%2Dsparse+transformers+have+interpretable+circuits%2E+arXiv+preprint+arXiv%3A2511%2E13653%2E+%E2%80%94+%5BarXiv%3A2511%2E1365)
82. Atticus Geiger, Duligur Ibeling, Amir Zur, Maheep Chaudhary, Sonakshi Chauhan, Jing Huang, Aryaman Arora, Zhengxuan Wu, Noah Goodman, Christopher Potts, and 1 others. 2025. Causal abstraction: A theoretical foundation for mechanistic interpretability. Journal of Machine Learning Research, 26(83):1–6… [find](https://scholar.google.com/scholar?q=Atticus+Geiger%2C+Duligur+Ibeling%2C+Amir+Zur%2C+Maheep+Chaudhary%2C+Sonakshi+Chauhan%2C+Jing+Huang%2C+Aryaman+Arora%2C+Zhengxuan+Wu%2C+Noah+Goodman%2C+Christopher+Potts%2C+and+1+others%2E+2025%2E+Causal+abstraction%3A+A+theor)
83. Mor Geva, Jasmijn Bastings, Katja Filippova, and Amir Globerson. 2023. Dissecting recall of factual associations in auto-regressive language models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 12216–12235, Singapore. Association for Computational… [find](https://scholar.google.com/scholar?q=Mor+Geva%2C+Jasmijn+Bastings%2C+Katja+Filippova%2C+and+Amir+Globerson%2E+2023%2E+Dissecting+recall+of+factual+associations+in+auto%2Dregressive+language+models%2E+In+Proceedings+of+the+2023+Conference+on+Empirical+)
84. Mor Geva, Avi Caciularu, Kevin Wang, and Yoav Goldberg. 2022. [Transformer feed-forward layers build predictions by promoting concepts in the vocabulary space](https://md.research.elamur.ai/free/en/papers/geva-2022-transformer-feed-forward-layers-build/). In Proceedings of the 2022 conference on empirical methods in natural language processing, pages 30–45.
85. Mor Geva, Roei Schuster, Jonathan Berant, and Omer Levy. 2021. Transformer feed-forward layers are key-value memories. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5484–5495, Online and Punta Cana, Dominican Republic. Association for Computational… [find](https://scholar.google.com/scholar?q=Mor+Geva%2C+Roei+Schuster%2C+Jonathan+Berant%2C+and+Omer+Levy%2E+2021%2E+Transformer+feed%2Dforward+layers+are+key%2Dvalue+memories%2E+In+Proceedings+of+the+2021+Conference+on+Empirical+Methods+in+Natural+Language+Pr)
86. Asma Ghandeharioun, Ann Yuan, Marius Guerard, Emily Reif, Michael A. Lepori, and Lucas Dixon. 2024. Who's asking? user personas and the mechanics of latent misalignment. In Advances in Neural Information Processing Systems, volume 37, pages 125967–126003. Curran Associates, Inc. [find](https://scholar.google.com/scholar?q=Asma+Ghandeharioun%2C+Ann+Yuan%2C+Marius+Guerard%2C+Emily+Reif%2C+Michael+A%2E+Lepori%2C+and+Lucas+Dixon%2E+2024%2E+Who%27s+asking%3F+user+personas+and+the+mechanics+of+latent+misalignment%2E+In+Advances+in+Neural+Informat)
87. Davide Ghilardi, Federico Belotti, and Marco Molinari. 2024. Efficient training of sparse autoencoders for large language models via layer groups. arXiv preprint arXiv:2410.21508. — [arXiv:2410.21508](https://arxiv.org/abs/2410.21508) [find](https://scholar.google.com/scholar?q=Davide+Ghilardi%2C+Federico+Belotti%2C+and+Marco+Molinari%2E+2024%2E+Efficient+training+of+sparse+autoencoders+for+large+language+models+via+layer+groups%2E+arXiv+preprint+arXiv%3A2410%2E21508%2E+%E2%80%94+%5BarXiv%3A2410%2E21508%5D)
88. Nicholas Goldowsky-Dill, Chris MacLeod, Lucas Sato, and Aryaman Arora. 2023. [Localizing model behavior with path patching](https://md.research.elamur.ai/free/en/papers/goldowsky-dill-2023-localizing-model-behavior-path-patching/). arXiv preprint arXiv:2304.05969.
89. Agam Goyal, Vedant Rathi, William Yeh, Yian Wang, Yuen Chen, and Hari Sundaram. 2025. Breaking Bad Tokens: Detoxification of LLMs Using Sparse Autoencoders. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 12702–12720, Suzhou, China. Association for Co… [find](https://scholar.google.com/scholar?q=Agam+Goyal%2C+Vedant+Rathi%2C+William+Yeh%2C+Yian+Wang%2C+Yuen+Chen%2C+and+Hari+Sundaram%2E+2025%2E+Breaking+Bad+Tokens%3A+Detoxification+of+LLMs+Using+Sparse+Autoencoders%2E+In+Proceedings+of+the+2025+Conference+on+Em)
90. Oliver Gruber and Thomas Goschke. 2004. Executive control emerging from dynamic interactions between brain systems mediating language, working memory and attentional processes. Acta psychologica, 115(2-3):105–121. [find](https://scholar.google.com/scholar?q=Oliver+Gruber+and+Thomas+Goschke%2E+2004%2E+Executive+control+emerging+from+dynamic+interactions+between+brain+systems+mediating+language%2C+working+memory+and+attentional+processes%2E+Acta+psychologica%2C+115%28)
91. Aleksandra Gruszka and Gerald Matthews. 2010. Handbook of individual differences in cognition: Attention, memory, and executive control. Springer. [find](https://scholar.google.com/scholar?q=Aleksandra+Gruszka+and+Gerald+Matthews%2E+2010%2E+Handbook+of+individual+differences+in+cognition%3A+Attention%2C+memory%2C+and+executive+control%2E+Springer%2E)
92. Xin Guan, PeiHsin Lin, Zekun Wu, Ze Wang, Ruibo Zhang, Emre Kazim, and Adriano Koshiyama. 2025. MPF: Aligning and Debiasing Language Models post Deployment via Multi Perspective Fusion. Preprint, arXiv:2507.02595. [find](https://scholar.google.com/scholar?q=Xin+Guan%2C+PeiHsin+Lin%2C+Zekun+Wu%2C+Ze+Wang%2C+Ruibo+Zhang%2C+Emre+Kazim%2C+and+Adriano+Koshiyama%2E+2025%2E+MPF%3A+Aligning+and+Debiasing+Language+Models+post+Deployment+via+Multi+Perspective+Fusion%2E+Preprint%2C+arXi)
93. Zhiyu Guo, Hidetaka Kamigaito, and Taro Watanabe. 2024. Attention score is not all you need for token importance indicator in KV cache reduction: Value also matters. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 21158–21166, Miami, Florida, USA. Ass… [find](https://scholar.google.com/scholar?q=Zhiyu+Guo%2C+Hidetaka+Kamigaito%2C+and+Taro+Watanabe%2E+2024%2E+Attention+score+is+not+all+you+need+for+token+importance+indicator+in+KV+cache+reduction%3A+Value+also+matters%2E+In+Proceedings+of+the+2024+Confere)
94. Zhengfu He, Xuyang Ge, Qiong Tang, Tianxiang Sun, Qinyuan Cheng, and Xipeng Qiu. 2024b. Dictionary learning improves patch-free circuit discovery in mechanistic interpretability: A case study on othello-gpt. arXiv preprint arXiv:2402.12201. — [arXiv:2402.12201](https://arxiv.org/abs/2402.12201) [find](https://scholar.google.com/scholar?q=Zhengfu+He%2C+Xuyang+Ge%2C+Qiong+Tang%2C+Tianxiang+Sun%2C+Qinyuan+Cheng%2C+and+Xipeng+Qiu%2E+2024b%2E+Dictionary+learning+improves+patch%2Dfree+circuit+discovery+in+mechanistic+interpretability%3A+A+case+study+on+othel)
95. Yoav Gur-Arieh, Roy Mayan, Chen Agassy, Atticus Geiger, and Mor Geva. 2025. [Enhancing automated interpretability with output-centric feature descriptions](https://md.research.elamur.ai/free/en/papers/arieh-2025-enhancing-automated-interpretability-output-centric/). arXiv preprint arXiv:2501.08319.
96. Zhengfu He, Wentao Shu, Xuyang Ge, Lingjie Chen, Junxuan Wang, Yunhua Zhou, Frances Liu, Qipeng Guo, Xuanjing Huang, Zuxuan Wu, and 1 others. 2024c. [Llama scope: Extracting millions of features from llama-3.1-8b with sparse autoencoders](https://md.research.elamur.ai/free/en/papers/he-2024-llama-scope-extracting-millions-features/). arXiv preprint arXiv:2410.20526.
97. Daniil Gurgurov, Katharina Trinley, Yusser Al Ghussin, Tanja Baeumel, Josef van Genabith, and Simon Ostermann. 2025a. Language arithmetics: Towards systematic language neuron identification and manipulation. Preprint, arXiv:2507.22608. [find](https://scholar.google.com/scholar?q=Daniil+Gurgurov%2C+Katharina+Trinley%2C+Yusser+Al+Ghussin%2C+Tanja+Baeumel%2C+Josef+van+Genabith%2C+and+Simon+Ostermann%2E+2025a%2E+Language+arithmetics%3A+Towards+systematic+language+neuron+identification+and+manipu)
98. Zirui He, Mingyu Jin, Bo Shen, Ali Payani, Yongfeng Zhang, and Mengnan Du. 2025a. Sae-ssv: Supervised steering in sparse representation spaces for reliable control of language models. arXiv preprint arXiv:2505.16188. [find](https://scholar.google.com/scholar?q=Zirui+He%2C+Mingyu+Jin%2C+Bo+Shen%2C+Ali+Payani%2C+Yongfeng+Zhang%2C+and+Mengnan+Du%2E+2025a%2E+Sae%2Dssv%3A+Supervised+steering+in+sparse+representation+spaces+for+reliable+control+of+language+models%2E+arXiv+preprint+a)
99. Daniil Gurgurov, Josef van Genabith, and Simon Ostermann. 2025b. Sparse subnetwork enhancement for underrepresented languages in large language models. Preprint, arXiv:2510.13580. — [arXiv:2510.13580](https://arxiv.org/abs/2510.13580) [find](https://scholar.google.com/scholar?q=Daniil+Gurgurov%2C+Josef+van+Genabith%2C+and+Simon+Ostermann%2E+2025b%2E+Sparse+subnetwork+enhancement+for+underrepresented+languages+in+large+language+models%2E+Preprint%2C+arXiv%3A2510%2E13580%2E+%E2%80%94+%5BarXiv%3A2510%2E13580%5D)
100. Zirui He, Haiyan Zhao, Yiran Qiao, Fan Yang, Ali Payani, Jing Ma, and Mengnan Du. 2025b. Saif: A sparse autoencoder framework for interpreting and steering instruction following of language models. arXiv preprint arXiv:2502.11356. — [arXiv:2502.11356](https://arxiv.org/abs/2502.11356) [find](https://scholar.google.com/scholar?q=Zirui+He%2C+Haiyan+Zhao%2C+Yiran+Qiao%2C+Fan+Yang%2C+Ali+Payani%2C+Jing+Ma%2C+and+Mengnan+Du%2E+2025b%2E+Saif%3A+A+sparse+autoencoder+framework+for+interpreting+and+steering+instruction+following+of+language+models%2E+ar)
101. Tal Haklay, Hadas Orgad, David Bau, Aaron Mueller, and Yonatan Belinkov. 2025. Position-aware automatic circuit discovery. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2025, Vienna, Austria, July 27 - August 1, 2025, pages 27… [find](https://scholar.google.com/scholar?q=Tal+Haklay%2C+Hadas+Orgad%2C+David+Bau%2C+Aaron+Mueller%2C+and+Yonatan+Belinkov%2E+2025%2E+Position%2Daware+automatic+circuit+discovery%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the+Association+for+Computationa)
102. Gunmay Handa, Zekun Wu, Adriano Koshiyama, and Philip Colin Treleaven. 2025. Personality as a probe for LLM evaluation: Method trade-offs and downstream effects. In NeurIPS 2025 Workshop on Evaluating the Evolving LLM Lifecycle: Benchmarks, Emergent Abilities, and Scaling. [find](https://scholar.google.com/scholar?q=Gunmay+Handa%2C+Zekun+Wu%2C+Adriano+Koshiyama%2C+and+Philip+Colin+Treleaven%2E+2025%2E+Personality+as+a+probe+for+LLM+evaluation%3A+Method+trade%2Doffs+and+downstream+effects%2E+In+NeurIPS+2025+Workshop+on+Evaluating)
103. Michael Hanna, Ollie Liu, and Alexandre Variengien. 2023. [How does GPT-2 compute greater-than?: Interpreting mathematical abilities in a pre-trained language model](https://md.research.elamur.ai/free/en/papers/hanna-2023-gpt-2-compute-greater-than/). In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023,…
104. Michael Hanna, Sandro Pezzelle, and Yonatan Belinkov. 2024. Have faith in faithfulness: Going beyond circuit overlap when finding model mechanisms. In First Conference on Language Modeling. [find](https://scholar.google.com/scholar?q=Michael+Hanna%2C+Sandro+Pezzelle%2C+and+Yonatan+Belinkov%2E+2024%2E+Have+faith+in+faithfulness%3A+Going+beyond+circuit+overlap+when+finding+model+mechanisms%2E+In+First+Conference+on+Language+Modeling%2E)
105. Michael Hanna, Mateusz Piotrowski, Jack Lindsey, and Emmanuel Ameisen. 2025. Circuit-tracer: A new library for finding feature circuits. In Proceedings of the 8th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 239– 249, Suzhou, China. Association for Computational Li… [find](https://scholar.google.com/scholar?q=Michael+Hanna%2C+Mateusz+Piotrowski%2C+Jack+Lindsey%2C+and+Emmanuel+Ameisen%2E+2025%2E+Circuit%2Dtracer%3A+A+new+library+for+finding+feature+circuits%2E+In+Proceedings+of+the+8th+BlackboxNLP+Workshop%3A+Analyzing+and+I)
106. Yefei He, Luoming Zhang, Weijia Wu, Jing Liu, Hong Zhou, and Bohan Zhuang. 2024a. Zipcache: Accurate and efficient kv cache quantization with salient token identification. Advances in Neural Information Processing Systems, 37:68287–68307. [find](https://scholar.google.com/scholar?q=Yefei+He%2C+Luoming+Zhang%2C+Weijia+Wu%2C+Jing+Liu%2C+Hong+Zhou%2C+and+Bohan+Zhuang%2E+2024a%2E+Zipcache%3A+Accurate+and+efficient+kv+cache+quantization+with+salient+token+identification%2E+Advances+in+Neural+Informati)
107. Amr Hegazy, Mostafa Elhoushi, and Amr Alanwar. 2025. Guiding giants: Lightweight controllers for weighted activation steering in llms. Preprint, arXiv:2505.20309. [find](https://scholar.google.com/scholar?q=Amr+Hegazy%2C+Mostafa+Elhoushi%2C+and+Amr+Alanwar%2E+2025%2E+Guiding+giants%3A+Lightweight+controllers+for+weighted+activation+steering+in+llms%2E+Preprint%2C+arXiv%3A2505%2E20309%2E)
108. Musashi Hinck, Carolin Holtermann, Matthew Lyle Olson, Florian Schneider, Sungduk Yu, Anahita Bhiwandiwalla, Anne Lauscher, Shao-Yen Tseng, and Vasudev Lal. 2024. [Why do LLaVA vision-language models reply to images in English?](https://md.research.elamur.ai/free/en/papers/hinck-2024-why-do-llava-vision-language/) In Findings of the Association for Computational Linguistics: EMNLP 2024,…
109. Yihuai Hong, Meng Cao, Dian Zhou, Lei Yu, and Zhijing Jin. 2025. The reasoning-memorization interplay in language models is mediated by a single direction. In Findings of the Association for Computational Linguistics: ACL 2025, pages 21565–21585, Vienna, Austria. Association for Computational Lingui… [find](https://scholar.google.com/scholar?q=Yihuai+Hong%2C+Meng+Cao%2C+Dian+Zhou%2C+Lei+Yu%2C+and+Zhijing+Jin%2E+2025%2E+The+reasoning%2Dmemorization+interplay+in+language+models+is+mediated+by+a+single+direction%2E+In+Findings+of+the+Association+for+Computati)
110. Jesse Hoogland, George Wang, Matthew Farrugia-Roberts, Liam Carroll, Susan Wei, and Daniel Murfet. 2024. The developmental landscape of in-context learning, 2024. URL https://arxiv.org/abs/2402.02364. — [arXiv:2402.02364](https://arxiv.org/abs/2402.02364) [↗](https://arxiv.org/abs/2402.02364)
111. Cheng-Hsun Hsueh, Paul Kuo-Ming Huang, Tzu-Han Lin, Che-Wei Liao, Hung-Chieh Fang, Chao-Wei Huang, and Yun-Nung Chen. 2024. Editing the mind of giants: An in-depth exploration of pitfalls of knowledge editing in large language models. arXiv preprint arXiv:2406.01436. [find](https://scholar.google.com/scholar?q=Cheng%2DHsun+Hsueh%2C+Paul+Kuo%2DMing+Huang%2C+Tzu%2DHan+Lin%2C+Che%2DWei+Liao%2C+Hung%2DChieh+Fang%2C+Chao%2DWei+Huang%2C+and+Yun%2DNung+Chen%2E+2024%2E+Editing+the+mind+of+giants%3A+An+in%2Ddepth+exploration+of+pitfalls+of+knowledge)
112. Lijie Hu, Chenyang Ren, Zhengyu Hu, Hongbin Lin, Cheng-Long Wang, Hui Xiong, Jingfeng Zhang, and Di Wang. 2025. Editable concept bottleneck models. Preprint, arXiv:2405.15476. — [arXiv:2405.15476](https://arxiv.org/abs/2405.15476) [find](https://scholar.google.com/scholar?q=Lijie+Hu%2C+Chenyang+Ren%2C+Zhengyu+Hu%2C+Hongbin+Lin%2C+Cheng%2DLong+Wang%2C+Hui+Xiong%2C+Jingfeng+Zhang%2C+and+Di+Wang%2E+2025%2E+Editable+concept+bottleneck+models%2E+Preprint%2C+arXiv%3A2405%2E15476%2E+%E2%80%94+%5BarXiv%3A2405%2E15476%5D%28htt)
113. Haoming Huang, Yibo Yan, Jiahao Huo, Xin Zou, Xinfeng Li, Kun Wang, and Xuming Hu. 2025a. Pierce the mists, greet the sky: Decipher knowledge overshadowing via knowledge circuit analysis. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 15471– 15490, S… [find](https://scholar.google.com/scholar?q=Haoming+Huang%2C+Yibo+Yan%2C+Jiahao+Huo%2C+Xin+Zou%2C+Xinfeng+Li%2C+Kun+Wang%2C+and+Xuming+Hu%2E+2025a%2E+Pierce+the+mists%2C+greet+the+sky%3A+Decipher+knowledge+overshadowing+via+knowledge+circuit+analysis%2E+In+Proceedin)
114. Jing Huang, Junyi Tao, Thomas Icard, Diyi Yang, and Christopher Potts. 2025b. Internal causal mechanisms robustly predict language model out-of-distribution behaviors. Preprint, arXiv:2505.11770. [find](https://scholar.google.com/scholar?q=Jing+Huang%2C+Junyi+Tao%2C+Thomas+Icard%2C+Diyi+Yang%2C+and+Christopher+Potts%2E+2025b%2E+Internal+causal+mechanisms+robustly+predict+language+model+out%2Dof%2Ddistribution+behaviors%2E+Preprint%2C+arXiv%3A2505%2E11770%2E)
115. Yufei Huang, Shengding Hu, Xu Han, Zhiyuan Liu, and Maosong Sun. 2024. Unified view of grokking, double descent and emergent abilities: A comprehensive study on algorithm task. In First Conference on Language Modeling. [find](https://scholar.google.com/scholar?q=Yufei+Huang%2C+Shengding+Hu%2C+Xu+Han%2C+Zhiyuan+Liu%2C+and+Maosong+Sun%2E+2024%2E+Unified+view+of+grokking%2C+double+descent+and+emergent+abilities%3A+A+comprehensive+study+on+algorithm+task%2E+In+First+Conference+on+)
116. Robert Huben, Hoagy Cunningham, Logan Riggs Smith, Aidan Ewart, and Lee Sharkey. 2024. Sparse autoencoders find highly interpretable features in language models. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. OpenReview.net. [find](https://scholar.google.com/scholar?q=Robert+Huben%2C+Hoagy+Cunningham%2C+Logan+Riggs+Smith%2C+Aidan+Ewart%2C+and+Lee+Sharkey%2E+2024%2E+Sparse+autoencoders+find+highly+interpretable+features+in+language+models%2E+In+The+Twelfth+International+Conferenc)
117. Jiahao Huo, Yibo Yan, Boren Hu, Yutao Yue, and Xuming Hu. 2024. Mmneuron: Discovering neuron-level domain-specific interpretation in multimodal large language model. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 6801–6816. [find](https://scholar.google.com/scholar?q=Jiahao+Huo%2C+Yibo+Yan%2C+Boren+Hu%2C+Yutao+Yue%2C+and+Xuming+Hu%2E+2024%2E+Mmneuron%3A+Discovering+neuron%2Dlevel+domain%2Dspecific+interpretation+in+multimodal+large+language+model%2E+In+Proceedings+of+the+2024+Confere)
118. Bertram Højer, Oliver Jarvis, and Stefan Heinrich. 2025. Improving reasoning performance in large language models via representation engineering. Preprint, arXiv:2504.19483. — [arXiv:2504.19483](https://arxiv.org/abs/2504.19483) [find](https://scholar.google.com/scholar?q=Bertram+H%C3%B8jer%2C+Oliver+Jarvis%2C+and+Stefan+Heinrich%2E+2025%2E+Improving+reasoning+performance+in+large+language+models+via+representation+engineering%2E+Preprint%2C+arXiv%3A2504%2E19483%2E+%E2%80%94+%5BarXiv%3A2504%2E19483%5D%28https)
119. Aya Abdelsalam Ismail, Tuomas Oikarinen, Amy Wang, Julius Adebayo, Samuel Stanton, Taylor Joren, Joseph Kleinhenz, Allen Goodman, Héctor Corrada Bravo, Kyunghyun Cho, and 1 others. 2024. Concept bottleneck language models for protein design. arXiv preprint arXiv:2411.06090. [find](https://scholar.google.com/scholar?q=Aya+Abdelsalam+Ismail%2C+Tuomas+Oikarinen%2C+Amy+Wang%2C+Julius+Adebayo%2C+Samuel+Stanton%2C+Taylor+Joren%2C+Joseph+Kleinhenz%2C+Allen+Goodman%2C+H%C3%A9ctor+Corrada+Bravo%2C+Kyunghyun+Cho%2C+and+1+others%2E+2024%2E+Concept+bottl)
120. Farnoush Rezaei Jafari, Oliver Eberle, Ashkan Khakzar, and Neel Nanda. 2025. Relp: Faithful and efficient circuit discovery via relevance patching. CoRR, abs/2508.21258. — [arXiv:2508.21258](https://arxiv.org/abs/2508.21258) [find](https://scholar.google.com/scholar?q=Farnoush+Rezaei+Jafari%2C+Oliver+Eberle%2C+Ashkan+Khakzar%2C+and+Neel+Nanda%2E+2025%2E+Relp%3A+Faithful+and+efficient+circuit+discovery+via+relevance+patching%2E+CoRR%2C+abs%2F2508%2E21258%2E+%E2%80%94+%5BarXiv%3A2508%2E21258%5D%28https%3A%2F%2Fa)
121. Che Jiang, Biqing Qi, Xiangyu Hong, Dayuan Fu, Yang Cheng, Fandong Meng, Mo Yu, Bowen Zhou, and Jie Zhou. 2024a. On large language models’ hallucination with regard to known facts. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: H… [find](https://scholar.google.com/scholar?q=Che+Jiang%2C+Biqing+Qi%2C+Xiangyu+Hong%2C+Dayuan+Fu%2C+Yang+Cheng%2C+Fandong+Meng%2C+Mo+Yu%2C+Bowen+Zhou%2C+and+Jie+Zhou%2E+2024a%2E+On+large+language+models%E2%80%99+hallucination+with+regard+to+known+facts%2E+In+Proceedings+of+t)
122. Gangwei Jiang, Zhaoyi Li, Caigao Jiang, Siqiao Xue, Jun Zhou, Linqi Song, Defu Lian, and Ying Wei. 2024b. Interpretable catastrophic forgetting of large language model fine-tuning via instruction vector. arXiv e-prints, pages arXiv–2406. [find](https://scholar.google.com/scholar?q=Gangwei+Jiang%2C+Zhaoyi+Li%2C+Caigao+Jiang%2C+Siqiao+Xue%2C+Jun+Zhou%2C+Linqi+Song%2C+Defu+Lian%2C+and+Ying+Wei%2E+2024b%2E+Interpretable+catastrophic+forgetting+of+large+language+model+fine%2Dtuning+via+instruction+vect)
123. Gangwei Jiang, Zhaoyi Li, Defu Lian, and Ying Wei. 2024c. Refine large language model fine-tuning via instruction vector. arXiv preprint arXiv:2406.12227. — [arXiv:2406.12227](https://arxiv.org/abs/2406.12227) [find](https://scholar.google.com/scholar?q=Gangwei+Jiang%2C+Zhaoyi+Li%2C+Defu+Lian%2C+and+Ying+Wei%2E+2024c%2E+Refine+large+language+model+fine%2Dtuning+via+instruction+vector%2E+arXiv+preprint+arXiv%3A2406%2E12227%2E+%E2%80%94+%5BarXiv%3A2406%2E12227%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F24)
124. Zhangqi Jiang, Junkai Chen, Beier Zhu, Tingjin Luo, Yankun Shen, and Xu Yang. 2025. Devils in middle layers of large vision-language models: Interpreting, detecting and mitigating object hallucinations via attention lens. In Proceedings of the Computer Vision and Pattern Recognition Conference, page… [find](https://scholar.google.com/scholar?q=Zhangqi+Jiang%2C+Junkai+Chen%2C+Beier+Zhu%2C+Tingjin+Luo%2C+Yankun+Shen%2C+and+Xu+Yang%2E+2025%2E+Devils+in+middle+layers+of+large+vision%2Dlanguage+models%3A+Interpreting%2C+detecting+and+mitigating+object+hallucination)
125. Mingyu Jin, Kai Mei, Wujiang Xu, Mingjie Sun, Ruixiang Tang, Mengnan Du, Zirui Liu, and Yongfeng Zhang. 2025a. Massive values in self-attention modules are the key to contextual knowledge understanding. In Forty-second International Conference on Machine Learning, ICML 2025, Vancouver, BC, Canada, J… [find](https://scholar.google.com/scholar?q=Mingyu+Jin%2C+Kai+Mei%2C+Wujiang+Xu%2C+Mingjie+Sun%2C+Ruixiang+Tang%2C+Mengnan+Du%2C+Zirui+Liu%2C+and+Yongfeng+Zhang%2E+2025a%2E+Massive+values+in+self%2Dattention+modules+are+the+key+to+contextual+knowledge+understandin)
126. Mingyu Jin, Qinkai Yu, Jingyuan Huang, Qingcheng Zeng, Zhenting Wang, Wenyue Hua, Haiyan Zhao, Kai Mei, Yanda Meng, Kaize Ding, Fan Yang, Mengnan Du, and Yongfeng Zhang. 2025b. [Exploring concept depth: How large language models acquire knowledge and concept at different layers?](https://md.research.elamur.ai/free/en/papers/jin-2025-exploring-concept-depth-how-large/) In Proceedings of the…
127. Zhuoran Jin, Pengfei Cao, Hongbang Yuan, Yubo Chen, Jiexin Xu, Huaijun Li, Xiaojian Jiang, Kang Liu, and Jun Zhao. 2024. Cutting off the head ends the conflict: A mechanism for interpreting and mitigating knowledge conflicts in language models. In Findings of the Association for Computational Lingui… [find](https://scholar.google.com/scholar?q=Zhuoran+Jin%2C+Pengfei+Cao%2C+Hongbang+Yuan%2C+Yubo+Chen%2C+Jiexin+Xu%2C+Huaijun+Li%2C+Xiaojian+Jiang%2C+Kang+Liu%2C+and+Jun+Zhao%2E+2024%2E+Cutting+off+the+head+ends+the+conflict%3A+A+mechanism+for+interpreting+and+mitiga)
128. Yi Jing, Zijun Yao, Hongzhu Guo, Lingxu Ran, Xiaozhi Wang, Lei Hou, and Juanzi Li. 2025. LinguaLens: Towards interpreting linguistic mechanisms of large language models via sparse auto-encoder. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 28220– 28… [find](https://scholar.google.com/scholar?q=Yi+Jing%2C+Zijun+Yao%2C+Hongzhu+Guo%2C+Lingxu+Ran%2C+Xiaozhi+Wang%2C+Lei+Hou%2C+and+Juanzi+Li%2E+2025%2E+LinguaLens%3A+Towards+interpreting+linguistic+mechanisms+of+large+language+models+via+sparse+auto%2Dencoder%2E+In+Pro)
129. Nitish Joshi, Javier Rando, Abulhair Saparov, Najoung Kim, and He He. 2024. Personas as a way to model truthfulness in language models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 6346–6359, Miami, Florida, USA. Association for Computational Lingu… [find](https://scholar.google.com/scholar?q=Nitish+Joshi%2C+Javier+Rando%2C+Abulhair+Saparov%2C+Najoung+Kim%2C+and+He+He%2E+2024%2E+Personas+as+a+way+to+model+truthfulness+in+language+models%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+Methods+in+Na)
130. Tianjie Ju, Zhenyu Shao, Bowen Wang, Yujia Chen, Zhuosheng Zhang, Hao Fei, Mong-Li Lee, Wynne Hsu, Sufeng Duan, and Gongshen Liu. 2025. Probing then editing response personality of large language models. In Second Conference on Language Modeling. [find](https://scholar.google.com/scholar?q=Tianjie+Ju%2C+Zhenyu+Shao%2C+Bowen+Wang%2C+Yujia+Chen%2C+Zhuosheng+Zhang%2C+Hao+Fei%2C+Mong%2DLi+Lee%2C+Wynne+Hsu%2C+Sufeng+Duan%2C+and+Gongshen+Liu%2E+2025%2E+Probing+then+editing+response+personality+of+large+language+mode)
131. Tianjie Ju, Weiwei Sun, Wei Du, Xinwei Yuan, Zhaochun Ren, and Gongshen Liu. 2024. How large language models encode context knowledge? A layer-wise probing study. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation, LREC/COLING 20… [find](https://scholar.google.com/scholar?q=Tianjie+Ju%2C+Weiwei+Sun%2C+Wei+Du%2C+Xinwei+Yuan%2C+Zhaochun+Ren%2C+and+Gongshen+Liu%2E+2024%2E+How+large+language+models+encode+context+knowledge%3F+A+layer%2Dwise+probing+study%2E+In+Proceedings+of+the+2024+Joint+Inte)
132. Cheongwoong Kang and Jaesik Choi. 2023. Impact of co-occurrence on factual knowledge of large language models. In Findings of the Association for Computational Linguistics: EMNLP 2023, Singapore, December 6-10, 2023, pages 7721–7735. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Cheongwoong+Kang+and+Jaesik+Choi%2E+2023%2E+Impact+of+co%2Doccurrence+on+factual+knowledge+of+large+language+models%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+EMNLP+2023%2C+Singapore%2C+Dece)
133. Subhash Kantamneni, Joshua Engels, Senthooran Rajamanoharan, Max Tegmark, and Neel Nanda. 2025. Are sparse autoencoders useful? A case study in sparse probing. In Forty-second International Conference on Machine Learning, ICML 2025, Vancouver, BC, Canada, July 13-19, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Subhash+Kantamneni%2C+Joshua+Engels%2C+Senthooran+Rajamanoharan%2C+Max+Tegmark%2C+and+Neel+Nanda%2E+2025%2E+Are+sparse+autoencoders+useful%3F+A+case+study+in+sparse+probing%2E+In+Forty%2Dsecond+International+Conference)
134. Amir Hossein Kargaran, Yihong Liu, François Yvon, and Hinrich Schütze. 2025. How programming concepts and neurons are shared in code language models. arXiv preprint arXiv:2506.01074. [find](https://scholar.google.com/scholar?q=Amir+Hossein+Kargaran%2C+Yihong+Liu%2C+Fran%C3%A7ois+Yvon%2C+and+Hinrich+Sch%C3%BCtze%2E+2025%2E+How+programming+concepts+and+neurons+are+shared+in+code+language+models%2E+arXiv+preprint+arXiv%3A2506%2E01074%2E)
135. Sheer Karny, Anthony Baez, and Pat Pataranutaporn. 2025. Neural transparency: Mechanistic interpretability interfaces for anticipating model behaviors for personalized ai. arXiv preprint arXiv:2511.00230. — [arXiv:2511.00230](https://arxiv.org/abs/2511.00230) [find](https://scholar.google.com/scholar?q=Sheer+Karny%2C+Anthony+Baez%2C+and+Pat+Pataranutaporn%2E+2025%2E+Neural+transparency%3A+Mechanistic+interpretability+interfaces+for+anticipating+model+behaviors+for+personalized+ai%2E+arXiv+preprint+arXiv%3A2511%2E00)
136. Adam Karvonen, Can Rager, Johnny Lin, Curt Tigges, Joseph Bloom, David Chanin, Yeu-Tong Lau, Eoin Farrell, Callum McDougall, Kola Ayonrinde, and 1 others. 2025. Saebench: A comprehensive benchmark for sparse autoencoders in language model interpretability. arXiv preprint arXiv:2503.09532. — [arXiv:2503.09532](https://arxiv.org/abs/2503.09532) [find](https://scholar.google.com/scholar?q=Adam+Karvonen%2C+Can+Rager%2C+Johnny+Lin%2C+Curt+Tigges%2C+Joseph+Bloom%2C+David+Chanin%2C+Yeu%2DTong+Lau%2C+Eoin+Farrell%2C+Callum+McDougall%2C+Kola+Ayonrinde%2C+and+1+others%2E+2025%2E+Saebench%3A+A+comprehensive+benchmark+for)
137. Aly M. Kassem, Zhuan Shi, Negar Rostamzadeh, and Golnoosh Farnadi. 2025. Reviving your MNEME: predicting the side effects of LLM unlearning and fine-tuning via sparse model diffing. CoRR, abs/2507.21084. [find](https://scholar.google.com/scholar?q=Aly+M%2E+Kassem%2C+Zhuan+Shi%2C+Negar+Rostamzadeh%2C+and+Golnoosh+Farnadi%2E+2025%2E+Reviving+your+MNEME%3A+predicting+the+side+effects+of+LLM+unlearning+and+fine%2Dtuning+via+sparse+model+diffing%2E+CoRR%2C+abs%2F2507%2E210)
138. Shahar Katz, Yonatan Belinkov, Mor Geva, and Lior Wolf. 2024. Backward lens: Projecting language model gradients into the vocabulary space. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA, November 12-16, 2024, pages 2390– 2422. A… [find](https://scholar.google.com/scholar?q=Shahar+Katz%2C+Yonatan+Belinkov%2C+Mor+Geva%2C+and+Lior+Wolf%2E+2024%2E+Backward+lens%3A+Projecting+language+model+gradients+into+the+vocabulary+space%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+Methods+i)
139. Junsol Kim, James Evans, and Aaron Schein. 2025b. Linear representations of political perspective emerge in large language models. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Junsol+Kim%2C+James+Evans%2C+and+Aaron+Schein%2E+2025b%2E+Linear+representations+of+political+perspective+emerge+in+large+language+models%2E+In+The+Thirteenth+International+Conference+on+Learning+Representation)
140. Junsol Kim, Shiyang Lai, Nino Scherrer, James Evans, and 1 others. 2026. Reasoning models generate societies of thought. arXiv preprint arXiv:2601.10825. — [arXiv:2601.10825](https://arxiv.org/abs/2601.10825) [find](https://scholar.google.com/scholar?q=Junsol+Kim%2C+Shiyang+Lai%2C+Nino+Scherrer%2C+James+Evans%2C+and+1+others%2E+2026%2E+Reasoning+models+generate+societies+of+thought%2E+arXiv+preprint+arXiv%3A2601%2E10825%2E+%E2%80%94+%5BarXiv%3A2601%2E10825%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F260)
141. Takeshi Kojima, Itsuki Okimura, Yusuke Iwasawa, Hitomi Yanaka, and Yutaka Matsuo. 2024. [On the multilingual ability of decoder-based pre-trained language models: Finding and controlling language-specific neurons](https://md.research.elamur.ai/free/en/papers/kojima-2024-multilingual-ability-decoder-based-pre/). In Proceedings of the 2024 Conference of the North American Chapter of the Association…
142. Wen Lai, Alexander Fraser, and Ivan Titov. 2025. Joint localization and activation editing for low-resource fine-tuning. In Forty-second International Conference on Machine Learning, ICML 2025, Vancouver, BC, Canada, July 13-19, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Wen+Lai%2C+Alexander+Fraser%2C+and+Ivan+Titov%2E+2025%2E+Joint+localization+and+activation+editing+for+low%2Dresource+fine%2Dtuning%2E+In+Forty%2Dsecond+International+Conference+on+Machine+Learning%2C+ICML+2025%2C+Vancou)
143. Filipe Laitenberger, Dawid Kopiczko, Cees G. M. Snoek, and Yuki M. Asano. 2025. What layers when: Learning to skip compute in llms with residual gates. Preprint, arXiv:2510.13876. — [arXiv:2510.13876](https://arxiv.org/abs/2510.13876) [find](https://scholar.google.com/scholar?q=Filipe+Laitenberger%2C+Dawid+Kopiczko%2C+Cees+G%2E+M%2E+Snoek%2C+and+Yuki+M%2E+Asano%2E+2025%2E+What+layers+when%3A+Learning+to+skip+compute+in+llms+with+residual+gates%2E+Preprint%2C+arXiv%3A2510%2E13876%2E+%E2%80%94+%5BarXiv%3A2510%2E13876%5D)
144. Tim Lawson and Laurence Aitchison. 2025. Learning to skip the middle layers of transformers. Preprint, arXiv:2506.21103. — [arXiv:2506.21103](https://arxiv.org/abs/2506.21103) [find](https://scholar.google.com/scholar?q=Tim+Lawson+and+Laurence+Aitchison%2E+2025%2E+Learning+to+skip+the+middle+layers+of+transformers%2E+Preprint%2C+arXiv%3A2506%2E21103%2E+%E2%80%94+%5BarXiv%3A2506%2E21103%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2506%2E21103%29)
145. Tim Lawson, Lucy Farnik, Conor Houghton, and Laurence Aitchison. 2024. Residual stream analysis with multi-layer saes. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Tim+Lawson%2C+Lucy+Farnik%2C+Conor+Houghton%2C+and+Laurence+Aitchison%2E+2024%2E+Residual+stream+analysis+with+multi%2Dlayer+saes%2E+In+The+Thirteenth+International+Conference+on+Learning+Representations%2E)
146. Andrew Lee, Xiaoyan Bai, Itamar Pres, Martin Wattenberg, Jonathan K. Kummerfeld, and Rada Mihalcea. 2024. [A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity](https://md.research.elamur.ai/free/en/papers/lee-2024-mechanistic-understanding-alignment-algorithms-case/). Preprint, arXiv:2401.01967.
147. Ihor Kendiukhov. 2025. A review of developmental interpretability in large language models. arXiv preprint arXiv:2508.15841. — [arXiv:2508.15841](https://arxiv.org/abs/2508.15841) [find](https://scholar.google.com/scholar?q=Ihor+Kendiukhov%2E+2025%2E+A+review+of+developmental+interpretability+in+large+language+models%2E+arXiv+preprint+arXiv%3A2508%2E15841%2E+%E2%80%94+%5BarXiv%3A2508%2E15841%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2508%2E15841%29)
148. Seongmin Lee, Aeree Cho, Grace C. Kim, ShengYun Peng, Mansi Phute, and Duen Horng Chau. 2025. Interpretation meets safety: A survey on interpretation methods and tools for improving LLM safety. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 21514–215… [find](https://scholar.google.com/scholar?q=Seongmin+Lee%2C+Aeree+Cho%2C+Grace+C%2E+Kim%2C+ShengYun+Peng%2C+Mansi+Phute%2C+and+Duen+Horng+Chau%2E+2025%2E+Interpretation+meets+safety%3A+A+survey+on+interpretation+methods+and+tools+for+improving+LLM+safety%2E+In+Pro)
149. Dmitrii Kharlapenko, Stepan Shabalin, Fazl Barez, Arthur Conmy, and Neel Nanda. 2025. Scaling sparse feature circuit finding for in-context learning. arXiv preprint arXiv:2504.13756. — [arXiv:2504.13756](https://arxiv.org/abs/2504.13756) [find](https://scholar.google.com/scholar?q=Dmitrii+Kharlapenko%2C+Stepan+Shabalin%2C+Fazl+Barez%2C+Arthur+Conmy%2C+and+Neel+Nanda%2E+2025%2E+Scaling+sparse+feature+circuit+finding+for+in%2Dcontext+learning%2E+arXiv+preprint+arXiv%3A2504%2E13756%2E+%E2%80%94+%5BarXiv%3A2504%2E137)
150. Lei Lei, Jie Gu, Xiaokang Ma, Chu Tang, Jingmin Chen, and Tong Xu. 2025. Generic token compression in multimodal large language models from an explainability perspective. Preprint, arXiv:2506.01097. — [arXiv:2506.01097](https://arxiv.org/abs/2506.01097) [find](https://scholar.google.com/scholar?q=Lei+Lei%2C+Jie+Gu%2C+Xiaokang+Ma%2C+Chu+Tang%2C+Jingmin+Chen%2C+and+Tong+Xu%2E+2025%2E+Generic+token+compression+in+multimodal+large+language+models+from+an+explainability+perspective%2E+Preprint%2C+arXiv%3A2506%2E01097%2E+%E2%80%94)
151. Jinyeong Kim, Seil Kang, Jiwoo Park, Junhyeok Kim, and Seong Jae Hwang. 2025a. Interpreting attention heads for image-to-text information flow in large vision–language models. In Mechanistic Interpretability Workshop at NeurIPS 2025. [find](https://scholar.google.com/scholar?q=Jinyeong+Kim%2C+Seil+Kang%2C+Jiwoo+Park%2C+Junhyeok+Kim%2C+and+Seong+Jae+Hwang%2E+2025a%2E+Interpreting+attention+heads+for+image%2Dto%2Dtext+information+flow+in+large+vision%E2%80%93language+models%2E+In+Mechanistic+Interpret)
152. Gaotang Li, Yuzhong Chen, and Hanghang Tong. 2025a. Taming knowledge conflicts in language models. In Forty-second International Conference on Machine Learning, ICML 2025, Vancouver, BC, Canada, July 13-19, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Gaotang+Li%2C+Yuzhong+Chen%2C+and+Hanghang+Tong%2E+2025a%2E+Taming+knowledge+conflicts+in+language+models%2E+In+Forty%2Dsecond+International+Conference+on+Machine+Learning%2C+ICML+2025%2C+Vancouver%2C+BC%2C+Canada%2C+July+)
153. Haoling Li, Xin Zhang, Xiao Liu, Yeyun Gong, Yifan Wang, Qi Chen, and Peng Cheng. 2025b. Enhancing large language model performance with gradient-based parameter selection. In AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelp… [find](https://scholar.google.com/scholar?q=Haoling+Li%2C+Xin+Zhang%2C+Xiao+Liu%2C+Yeyun+Gong%2C+Yifan+Wang%2C+Qi+Chen%2C+and+Peng+Cheng%2E+2025b%2E+Enhancing+large+language+model+performance+with+gradient%2Dbased+parameter+selection%2E+In+AAAI%2D25%2C+Sponsored+by+th)
154. Jiaming Li, Haoran Ye, Yukun Chen, Xinyue Li, Lei Zhang, Hamid Alinejad-Rokny, Jimmy Chih-Hsien Peng, and Min Yang. 2025c. Training superior sparse autoencoders for instruct models. arXiv preprint arXiv:2506.07691. — [arXiv:2506.07691](https://arxiv.org/abs/2506.07691) [find](https://scholar.google.com/scholar?q=Jiaming+Li%2C+Haoran+Ye%2C+Yukun+Chen%2C+Xinyue+Li%2C+Lei+Zhang%2C+Hamid+Alinejad%2DRokny%2C+Jimmy+Chih%2DHsien+Peng%2C+and+Min+Yang%2E+2025c%2E+Training+superior+sparse+autoencoders+for+instruct+models%2E+arXiv+preprint+arX)
155. Junnan Li, Dongxu Li, Silvio Savarese, and Steven C. H. Hoi. 2023. [BLIP-2: bootstrapping language-image pre-training with frozen image encoders and large language models](https://md.research.elamur.ai/free/en/papers/li-2023-blip-2-bootstrapping-language-image/). In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, Proceedings of Machine Learni…
156. Ruizhe Li, Chen Chen, Yuchen Hu, Yanjun Gao, Xi Wang, and Emine Yilmaz. 2026. Attributing response to context: A jensen-shannon divergence driven mechanistic study of context attribution in retrieval-augmented generation. In The Fourteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Ruizhe+Li%2C+Chen+Chen%2C+Yuchen+Hu%2C+Yanjun+Gao%2C+Xi+Wang%2C+and+Emine+Yilmaz%2E+2026%2E+Attributing+response+to+context%3A+A+jensen%2Dshannon+divergence+driven+mechanistic+study+of+context+attribution+in+retrieval%2D)
157. Ruizhe Li and Yanjun Gao. 2025. Anchored Answers: Unravelling Positional Bias in GPT-2’s Multiple-Choice Questions. Preprint, arXiv:2405.03205. [find](https://scholar.google.com/scholar?q=Ruizhe+Li+and+Yanjun+Gao%2E+2025%2E+Anchored+Answers%3A+Unravelling+Positional+Bias+in+GPT%2D2%E2%80%99s+Multiple%2DChoice+Questions%2E+Preprint%2C+arXiv%3A2405%2E03205%2E)
158. Shen Li, Liuyi Yao, Lan Zhang, and Yaliang Li. 2025d. Safety layers in aligned large language models: The key to LLM security. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Shen+Li%2C+Liuyi+Yao%2C+Lan+Zhang%2C+and+Yaliang+Li%2E+2025d%2E+Safety+layers+in+aligned+large+language+models%3A+The+key+to+LLM+security%2E+In+The+Thirteenth+International+Conference+on+Learning+Representations%2E)
159. Xuying Li, Zhuo Li, Yuji Kosuga, Yasuhiro Yoshida, and Victor Bian. 2024. Precision Knowledge Editing: Enhancing Safety in Large Language Models. Preprint, arXiv:2410.03772. — [arXiv:2410.03772](https://arxiv.org/abs/2410.03772) [find](https://scholar.google.com/scholar?q=Xuying+Li%2C+Zhuo+Li%2C+Yuji+Kosuga%2C+Yasuhiro+Yoshida%2C+and+Victor+Bian%2E+2024%2E+Precision+Knowledge+Editing%3A+Enhancing+Safety+in+Large+Language+Models%2E+Preprint%2C+arXiv%3A2410%2E03772%2E+%E2%80%94+%5BarXiv%3A2410%2E03772%5D%28https)
160. Yueyan Li, Wenhao Gao, Caixia Yuan, and Xiaojie Wang. 2025e. Fine-tuning is subgraph search: A new lens on learning dynamics. Preprint, arXiv:2502.06106. — [arXiv:2502.06106](https://arxiv.org/abs/2502.06106) [find](https://scholar.google.com/scholar?q=Yueyan+Li%2C+Wenhao+Gao%2C+Caixia+Yuan%2C+and+Xiaojie+Wang%2E+2025e%2E+Fine%2Dtuning+is+subgraph+search%3A+A+new+lens+on+learning+dynamics%2E+Preprint%2C+arXiv%3A2502%2E06106%2E+%E2%80%94+%5BarXiv%3A2502%2E06106%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F250)
161. Zhong-Zhi Li, Duzhen Zhang, Ming-Liang Zhang, Jiaxin Zhang, Zengyan Liu, Yuxuan Yao, Haotian Xu, Junhao Zheng, Pei-Jie Wang, Xiuyi Chen, and 1 others. 2025f. [From system 1 to system 2: A survey of reasoning large language models](https://md.research.elamur.ai/free/en/papers/li-2025-system-1-system-2-survey/). arXiv preprint arXiv:2502.17419.
162. Zihao Li, Xu Wang, Yuzhe Yang, Ziyu Yao, Haoyi Xiong, and Mengnan Du. 2025g. [Feature extraction and steering for enhanced chain-of-thought reasoning in language models](https://md.research.elamur.ai/free/en/papers/li-2025-feature-extraction-steering-enhanced-chain/). In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 10893–10913, Suzhou, China. Associ…
163. Ziyue Li, Chenrui Fan, and Tianyi Zhou. 2025h. Where to find grokking in llm pretraining? monitor memorization-to-generalization without test. arXiv preprint arXiv:2506.21551. — [arXiv:2506.21551](https://arxiv.org/abs/2506.21551) [find](https://scholar.google.com/scholar?q=Ziyue+Li%2C+Chenrui+Fan%2C+and+Tianyi+Zhou%2E+2025h%2E+Where+to+find+grokking+in+llm+pretraining%3F+monitor+memorization%2Dto%2Dgeneralization+without+test%2E+arXiv+preprint+arXiv%3A2506%2E21551%2E+%E2%80%94+%5BarXiv%3A2506%2E21551%5D%28htt)
164. Xiao Liang, Zhong-Zhi Li, Yeyun Gong, Yang Wang, Hengyuan Zhang, Yelong Shen, Ying Nian Wu, and Weizhu Chen. 2025. Sws: Self-aware weakness-driven problem synthesis in reinforcement learning for llm reasoning. arXiv preprint arXiv:2506.08989. — [arXiv:2506.08989](https://arxiv.org/abs/2506.08989) [find](https://scholar.google.com/scholar?q=Xiao+Liang%2C+Zhong%2DZhi+Li%2C+Yeyun+Gong%2C+Yang+Wang%2C+Hengyuan+Zhang%2C+Yelong+Shen%2C+Ying+Nian+Wu%2C+and+Weizhu+Chen%2E+2025%2E+Sws%3A+Self%2Daware+weakness%2Ddriven+problem+synthesis+in+reinforcement+learning+for+llm+r)
165. Xiao Liang, Zhong-Zhi Li, Zhenghao Lin, Eric Hancheng Jiang, Hengyuan Zhang, Yelong Shen, Kai-Wei Chang, Ying Nian Wu, Yeyun Gong, and Weizhu Chen. 2026. Training llms for divide-and-conquer reasoning elevates test-time scalability. arXiv preprint arXiv:2602.02477. [find](https://scholar.google.com/scholar?q=Xiao+Liang%2C+Zhong%2DZhi+Li%2C+Zhenghao+Lin%2C+Eric+Hancheng+Jiang%2C+Hengyuan+Zhang%2C+Yelong+Shen%2C+Kai%2DWei+Chang%2C+Ying+Nian+Wu%2C+Yeyun+Gong%2C+and+Weizhu+Chen%2E+2026%2E+Training+llms+for+divide%2Dand%2Dconquer+reasoning)
166. Jindřich Libovický, Rudolf Rosa, and Alexander Fraser. 2020. [On the language neutrality of pre-trained multilingual representations](https://md.research.elamur.ai/free/en/papers/libovicky-2020-language-neutrality-pre-trained-multilingual/). In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1663–1674, Online. Association for Computational Linguistics.
167. Tom Lieberum, Senthooran Rajamanoharan, Arthur Conmy, Lewis Smith, Nicolas Sonnerat, Vikrant Varma, Janos Kramar, Anca Dragan, Rohin Shah, and Neel Nanda. 2024. [Gemma scope: Open sparse autoencoders everywhere all at once on gemma 2](https://md.research.elamur.ai/free/en/papers/lieberum-2024-gemma-scope-open-sparse-autoencoders/). In Proceedings of the 7th BlackboxNLP Workshop: Analyzing and Inte…
168. Ji Lin, Jiaming Tang, Haotian Tang, Shang Yang, Wei-Ming Chen, Wei-Chen Wang, Guangxuan Xiao, Xingyu Dang, Chuang Gan, and Song Han. 2024. Awq: Activation-aware weight quantization for on-device llm compression and acceleration. Proceedings of machine learning and systems, 6:87–100. [find](https://scholar.google.com/scholar?q=Ji+Lin%2C+Jiaming+Tang%2C+Haotian+Tang%2C+Shang+Yang%2C+Wei%2DMing+Chen%2C+Wei%2DChen+Wang%2C+Guangxuan+Xiao%2C+Xingyu+Dang%2C+Chuang+Gan%2C+and+Song+Han%2E+2024%2E+Awq%3A+Activation%2Daware+weight+quantization+for+on%2Ddevice+llm+c)
169. Zihao Lin, Samyadeep Basu, Mohammad Beigi, Varun Manjunatha, Ryan A Rossi, Zichao Wang, Yufan Zhou, Sriram Balasubramanian, Arman Zarei, Keivan Rezaei, and 1 others. 2025. A survey on mechanistic interpretability for multi-modal foundation models. arXiv preprint arXiv:2502.17516. — [arXiv:2502.17516](https://arxiv.org/abs/2502.17516) [find](https://scholar.google.com/scholar?q=Zihao+Lin%2C+Samyadeep+Basu%2C+Mohammad+Beigi%2C+Varun+Manjunatha%2C+Ryan+A+Rossi%2C+Zichao+Wang%2C+Yufan+Zhou%2C+Sriram+Balasubramanian%2C+Arman+Zarei%2C+Keivan+Rezaei%2C+and+1+others%2E+2025%2E+A+survey+on+mechanistic+inte)
170. Jack Lindsey, Wes Gurnee, Emmanuel Ameisen, Brian Chen, Adam Pearce, Nicholas L. Turner, Craig Citro, David Abrahams, Shan Carter, Basil Hosmer, Jonathan Marcus, Michael Sklar, Adly Templeton, Trenton Bricken, Callum McDougall, Hoagy Cunningham, Thomas Henighan, Adam Jermyn, Andy Jones, and 8 others… [find](https://scholar.google.com/scholar?q=Jack+Lindsey%2C+Wes+Gurnee%2C+Emmanuel+Ameisen%2C+Brian+Chen%2C+Adam+Pearce%2C+Nicholas+L%2E+Turner%2C+Craig+Citro%2C+David+Abrahams%2C+Shan+Carter%2C+Basil+Hosmer%2C+Jonathan+Marcus%2C+Michael+Sklar%2C+Adly+Templeton%2C+Trenton)
171. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023a. [Visual instruction tuning](https://md.research.elamur.ai/free/en/papers/liu-2023-visual-instruction-tuning/). In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023.
172. Qi Liu, Haozhe Duan, Jiaxin Mao, and Ji-Rong Wen. 2025a. How do large language models understand relevance? a mechanistic interpretability perspective. ACM Transactions on Information Systems. [find](https://scholar.google.com/scholar?q=Qi+Liu%2C+Haozhe+Duan%2C+Jiaxin+Mao%2C+and+Ji%2DRong+Wen%2E+2025a%2E+How+do+large+language+models+understand+relevance%3F+a+mechanistic+interpretability+perspective%2E+ACM+Transactions+on+Information+Systems%2E)
173. Sheng Liu, Tianlang Chen, Pan Lu, Haotian Ye, Yizheng Chen, Lei Xing, and James Zou. 2025b. Fractional reasoning via latent steering vectors improves inference time compute. Preprint, arXiv:2506.15882. — [arXiv:2506.15882](https://arxiv.org/abs/2506.15882) [find](https://scholar.google.com/scholar?q=Sheng+Liu%2C+Tianlang+Chen%2C+Pan+Lu%2C+Haotian+Ye%2C+Yizheng+Chen%2C+Lei+Xing%2C+and+James+Zou%2E+2025b%2E+Fractional+reasoning+via+latent+steering+vectors+improves+inference+time+compute%2E+Preprint%2C+arXiv%3A2506%2E15882)
174. Shuqi Liu, Han Wu, Bowei He, Xiongwei Han, Mingxuan Yuan, and Linqi Song. 2025c. Sens-merging: Sensitivity-guided parameter balancing for merging large language models. In Findings of the Association for Computational Linguistics, ACL 2025, Vienna, Austria, July 27 - August 1, 2025, pages 19243– 192… [find](https://scholar.google.com/scholar?q=Shuqi+Liu%2C+Han+Wu%2C+Bowei+He%2C+Xiongwei+Han%2C+Mingxuan+Yuan%2C+and+Linqi+Song%2E+2025c%2E+Sens%2Dmerging%3A+Sensitivity%2Dguided+parameter+balancing+for+merging+large+language+models%2E+In+Findings+of+the+Association+)
175. Tianlin Liu, Mathieu Blondel, Carlos Riquelme Ruiz, and Joan Puigcerver. 2024a. Routers in vision mixture of experts: An empirical study. [find](https://scholar.google.com/scholar?q=Tianlin+Liu%2C+Mathieu+Blondel%2C+Carlos+Riquelme+Ruiz%2C+and+Joan+Puigcerver%2E+2024a%2E+Routers+in+vision+mixture+of+experts%3A+An+empirical+study%2E)
176. Trans. Mach. Learn. Res., 2024. [find](https://scholar.google.com/scholar?q=Trans%2E+Mach%2E+Learn%2E+Res%2E%2C+2024%2E)
177. Weize Liu, Yinlong Xu, Hongxia Xu, Jintai Chen, Xuming Hu, and Jian Wu. 2024b. Unraveling babel: Exploring multilingual activation patterns within large language models. arXiv. [find](https://scholar.google.com/scholar?q=Weize+Liu%2C+Yinlong+Xu%2C+Hongxia+Xu%2C+Jintai+Chen%2C+Xuming+Hu%2C+and+Jian+Wu%2E+2024b%2E+Unraveling+babel%3A+Exploring+multilingual+activation+patterns+within+large+language+models%2E+arXiv%2E)
178. Yan Liu, Yu Liu, Xiaokang Chen, Pin-Yu Chen, Daoguang Zan, Min-Yen Kan, and Tsung-Yi Ho. 2024c. The Devil is in the Neurons: Interpreting and Mitigating Social Biases in Pre-trained Language Models. Preprint, arXiv:2406.10130. — [arXiv:2406.10130](https://arxiv.org/abs/2406.10130) [find](https://scholar.google.com/scholar?q=Yan+Liu%2C+Yu+Liu%2C+Xiaokang+Chen%2C+Pin%2DYu+Chen%2C+Daoguang+Zan%2C+Min%2DYen+Kan%2C+and+Tsung%2DYi+Ho%2E+2024c%2E+The+Devil+is+in+the+Neurons%3A+Interpreting+and+Mitigating+Social+Biases+in+Pre%2Dtrained+Language+Models%2E+P)
179. Yihong Liu, Runsheng Chen, Lea Hirlimann, Ahmad Dawar Hakimi, Mingyang Wang, Amir Hossein Kargaran, Sascha Rothe, François Yvon, and Hinrich Schuetze. 2025d. On relation-specific neurons in large language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processi… [find](https://scholar.google.com/scholar?q=Yihong+Liu%2C+Runsheng+Chen%2C+Lea+Hirlimann%2C+Ahmad+Dawar+Hakimi%2C+Mingyang+Wang%2C+Amir+Hossein+Kargaran%2C+Sascha+Rothe%2C+Fran%C3%A7ois+Yvon%2C+and+Hinrich+Schuetze%2E+2025d%2E+On+relation%2Dspecific+neurons+in+large+lang)
180. Yihong Liu, Mingyang Wang, Amir Hossein Kargaran, Felicia Körner, Ercong Nie, Barbara Plank, François Yvon, and Hinrich Schuetze. 2025e. Tracing multilingual factual knowledge acquisition in pretraining. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 2121–2146, Suzho… [find](https://scholar.google.com/scholar?q=Yihong+Liu%2C+Mingyang+Wang%2C+Amir+Hossein+Kargaran%2C+Felicia+K%C3%B6rner%2C+Ercong+Nie%2C+Barbara+Plank%2C+Fran%C3%A7ois+Yvon%2C+and+Hinrich+Schuetze%2E+2025e%2E+Tracing+multilingual+factual+knowledge+acquisition+in+pretraini)
181. Ziming Liu, Eric J Michaud, and Max Tegmark. 2023b. Omnigrok: Grokking beyond algorithmic data. In The Eleventh International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Ziming+Liu%2C+Eric+J+Michaud%2C+and+Max+Tegmark%2E+2023b%2E+Omnigrok%3A+Grokking+beyond+algorithmic+data%2E+In+The+Eleventh+International+Conference+on+Learning+Representations%2E)
182. Haoyan Luo and Lucia Specia. 2024. From understanding to utilization: A survey on explainability for large language models. arXiv preprint arXiv:2401.12874. — [arXiv:2401.12874](https://arxiv.org/abs/2401.12874) [find](https://scholar.google.com/scholar?q=Haoyan+Luo+and+Lucia+Specia%2E+2024%2E+From+understanding+to+utilization%3A+A+survey+on+explainability+for+large+language+models%2E+arXiv+preprint+arXiv%3A2401%2E12874%2E+%E2%80%94+%5BarXiv%3A2401%2E12874%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F)
183. Ang Lv, Kaiyi Zhang, Yuhan Chen, Yulong Wang, Lifeng Liu, Ji-Rong Wen, Jian Xie, and Rui Yan. 2024. Interpreting key mechanisms of factual recall in transformer-based language models. CoRR, abs/2403.19521. — [arXiv:2403.19521](https://arxiv.org/abs/2403.19521) [find](https://scholar.google.com/scholar?q=Ang+Lv%2C+Kaiyi+Zhang%2C+Yuhan+Chen%2C+Yulong+Wang%2C+Lifeng+Liu%2C+Ji%2DRong+Wen%2C+Jian+Xie%2C+and+Rui+Yan%2E+2024%2E+Interpreting+key+mechanisms+of+factual+recall+in+transformer%2Dbased+language+models%2E+CoRR%2C+abs%2F2403%2E1)
184. Samuel Marks, Can Rager, Eric J. Michaud, Yonatan Belinkov, David Bau, and Aaron Mueller. 2025. Sparse feature circuits: Discovering and editing interpretable causal graphs in language models. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28,… [find](https://scholar.google.com/scholar?q=Samuel+Marks%2C+Can+Rager%2C+Eric+J%2E+Michaud%2C+Yonatan+Belinkov%2C+David+Bau%2C+and+Aaron+Mueller%2E+2025%2E+Sparse+feature+circuits%3A+Discovering+and+editing+interpretable+causal+graphs+in+language+models%2E+In+The+)
185. Callum Stuart McDougall, Arthur Conmy, Cody Rushing, Thomas McGrath, and Neel Nanda. 2024. Copy suppression: Comprehensively understanding a motif in language model attention heads. In Proceedings of the 7th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 337–363, Mia… [find](https://scholar.google.com/scholar?q=Callum+Stuart+McDougall%2C+Arthur+Conmy%2C+Cody+Rushing%2C+Thomas+McGrath%2C+and+Neel+Nanda%2E+2024%2E+Copy+suppression%3A+Comprehensively+understanding+a+motif+in+language+model+attention+heads%2E+In+Proceedings+of+)
186. Tianyi Men, Pengfei Cao, Zhuoran Jin, Yubo Chen, Kang Liu, and Jun Zhao. 2024. Unlocking the future: Exploring look-ahead planning mechanistic interpretability in large language models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 7713–7724. [find](https://scholar.google.com/scholar?q=Tianyi+Men%2C+Pengfei+Cao%2C+Zhuoran+Jin%2C+Yubo+Chen%2C+Kang+Liu%2C+and+Jun+Zhao%2E+2024%2E+Unlocking+the+future%3A+Exploring+look%2Dahead+planning+mechanistic+interpretability+in+large+language+models%2E+In+Proceedings)
187. Xin Men, Mingyu Xu, Qingyu Zhang, Qianhao Yuan, Bingning Wang, Hongyu Lin, Yaojie Lu, Xianpei Han, and Weipeng Chen. 2025. [ShortGPT: Layers in large language models are more redundant than you expect](https://md.research.elamur.ai/free/en/papers/men-2024-shortgpt-layers-large-language-models/). In Findings of the Association for Computational Linguistics: ACL 2025, pages 20192–20204, Vienna,…
188. Kevin Meng, David Bau, Alex Andonian, and Yonatan Belinkov. 2022. Locating and editing factual associations in GPT. In Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, November 28 - December 9,… [find](https://scholar.google.com/scholar?q=Kevin+Meng%2C+David+Bau%2C+Alex+Andonian%2C+and+Yonatan+Belinkov%2E+2022%2E+Locating+and+editing+factual+associations+in+GPT%2E+In+Advances+in+Neural+Information+Processing+Systems+35%3A+Annual+Conference+on+Neural)
189. Christina Lu, Jack Gallagher, Jonathan Michala, Kyle Fish, and Jack Lindsey. 2026. The assistant axis: Situating and stabilizing the default persona of language models. arXiv preprint arXiv:2601.10387. — [arXiv:2601.10387](https://arxiv.org/abs/2601.10387) [find](https://scholar.google.com/scholar?q=Christina+Lu%2C+Jack+Gallagher%2C+Jonathan+Michala%2C+Kyle+Fish%2C+and+Jack+Lindsey%2E+2026%2E+The+assistant+axis%3A+Situating+and+stabilizing+the+default+persona+of+language+models%2E+arXiv+preprint+arXiv%3A2601%2E10387)
190. Kevin Meng, Arnab Sen Sharma, Alex J Andonian, Yonatan Belinkov, and David Bau. 2023. Mass-editing memory in a transformer. In The Eleventh International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Kevin+Meng%2C+Arnab+Sen+Sharma%2C+Alex+J+Andonian%2C+Yonatan+Belinkov%2C+and+David+Bau%2E+2023%2E+Mass%2Dediting+memory+in+a+transformer%2E+In+The+Eleventh+International+Conference+on+Learning+Representations%2E)
191. Dawn Lu and Nina Rimsky. 2024. Investigating bias representations in llama 2 chat via activation steering. arXiv preprint arXiv:2402.00402. — [arXiv:2402.00402](https://arxiv.org/abs/2402.00402) [find](https://scholar.google.com/scholar?q=Dawn+Lu+and+Nina+Rimsky%2E+2024%2E+Investigating+bias+representations+in+llama+2+chat+via+activation+steering%2E+arXiv+preprint+arXiv%3A2402%2E00402%2E+%E2%80%94+%5BarXiv%3A2402%2E00402%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2402%2E00402%29)
192. Joseph Miller, Bilal Chughtai, and William Saunders. 2024. Transformer circuit faithfulness metrics are not robust. arXiv preprint arXiv:2407.08734. — [arXiv:2407.08734](https://arxiv.org/abs/2407.08734) [find](https://scholar.google.com/scholar?q=Joseph+Miller%2C+Bilal+Chughtai%2C+and+William+Saunders%2E+2024%2E+Transformer+circuit+faithfulness+metrics+are+not+robust%2E+arXiv+preprint+arXiv%3A2407%2E08734%2E+%E2%80%94+%5BarXiv%3A2407%2E08734%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2407%2E087)
193. Xudong Lu, Qi Liu, Yuhui Xu, Aojun Zhou, Siyuan Huang, Bo Zhang, Junchi Yan, and Hongsheng Li. 2024. [Not all experts are equal: Efficient expert pruning and skipping for mixture-of-experts large language models](https://md.research.elamur.ai/free/en/papers/lu-2024-not-all-experts-are-equal/). arXiv preprint arXiv:2402.14800.
194. Gouki Minegishi, Hiroki Furuta, Shohei Taniguchi, Yusuke Iwasawa, and Yutaka Matsuo. 2025. In-context meta learning induces multi-phase circuit emergence. In ICLR 2025 Workshop on Building Trust in Language Models and Applications. [find](https://scholar.google.com/scholar?q=Gouki+Minegishi%2C+Hiroki+Furuta%2C+Shohei+Taniguchi%2C+Yusuke+Iwasawa%2C+and+Yutaka+Matsuo%2E+2025%2E+In%2Dcontext+meta+learning+induces+multi%2Dphase+circuit+emergence%2E+In+ICLR+2025+Workshop+on+Building+Trust+in+La)
195. Soumen Kumar Mondal, Sayambhu Sen, Abhishek Singhania, and Preethi Jyothi. 2025. Language-specific neurons do not facilitate cross-lingual transfer. In The Sixth Workshop on Insights from Negative Results in NLP, pages 46–62, Albuquerque, New Mexico. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Soumen+Kumar+Mondal%2C+Sayambhu+Sen%2C+Abhishek+Singhania%2C+and+Preethi+Jyothi%2E+2025%2E+Language%2Dspecific+neurons+do+not+facilitate+cross%2Dlingual+transfer%2E+In+The+Sixth+Workshop+on+Insights+from+Negative+Res)
196. Clement Neo, Luke Ong, Philip Torr, Mor Geva, David Krueger, and Fazl Barez. 2025. Towards interpreting visual information processing in vision-language models. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net. [find](https://scholar.google.com/scholar?q=Clement+Neo%2C+Luke+Ong%2C+Philip+Torr%2C+Mor+Geva%2C+David+Krueger%2C+and+Fazl+Barez%2E+2025%2E+Towards+interpreting+visual+information+processing+in+vision%2Dlanguage+models%2E+In+The+Thirteenth+International+Confere)
197. John Jacob Brooke Morgan and Adam Raymond Gilliland. 1927. An introduction to psychology. Macmillan. [find](https://scholar.google.com/scholar?q=John+Jacob+Brooke+Morgan+and+Adam+Raymond+Gilliland%2E+1927%2E+An+introduction+to+psychology%2E+Macmillan%2E)
198. Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, and Mohit Bansal. 2025a. [Grains: Gradient-based attribution for inference-time steering of llms and vlms](https://md.research.elamur.ai/free/en/papers/nguyen-2026-grains-gradient-based-attribution-inference/). CoRR, abs/2507.18043.
199. Basel Mousi, Nadir Durrani, Fahim Dalvi, Majd Hawasly, and Ahmed Abdelali. 2024. Exploring alignment in shared cross-lingual spaces. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6326–6348, Bangkok, Thailand. Association for… [find](https://scholar.google.com/scholar?q=Basel+Mousi%2C+Nadir+Durrani%2C+Fahim+Dalvi%2C+Majd+Hawasly%2C+and+Ahmed+Abdelali%2E+2024%2E+Exploring+alignment+in+shared+cross%2Dlingual+spaces%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+Association+for+Co)
200. Anish Mudide, Joshua Engels, Eric J Michaud, Max Tegmark, and Christian Schroeder de Witt. 2024. Efficient dictionary learning with switch sparse autoencoders. arXiv preprint arXiv:2410.08201. — [arXiv:2410.08201](https://arxiv.org/abs/2410.08201) [find](https://scholar.google.com/scholar?q=Anish+Mudide%2C+Joshua+Engels%2C+Eric+J+Michaud%2C+Max+Tegmark%2C+and+Christian+Schroeder+de+Witt%2E+2024%2E+Efficient+dictionary+learning+with+switch+sparse+autoencoders%2E+arXiv+preprint+arXiv%3A2410%2E08201%2E+%E2%80%94+%5BarXi)
201. Aaron Mueller, Atticus Geiger, Sarah Wiegreffe, Dana Arad, Iván Arcuschin, Adam Belfki, Yik Siu Chan, Jaden Fried Fiotto-Kaufman, Tal Haklay, Michael Hanna, Jing Huang, Rohan Gupta, Yaniv Nikankin, Hadas Orgad, Nikhil Prakash, Anja Reusch, Aruna Sankaranarayanan, Shun Shao, Alessandro Stolfo, and 4… [find](https://scholar.google.com/scholar?q=Aaron+Mueller%2C+Atticus+Geiger%2C+Sarah+Wiegreffe%2C+Dana+Arad%2C+Iv%C3%A1n+Arcuschin%2C+Adam+Belfki%2C+Yik+Siu+Chan%2C+Jaden+Fried+Fiotto%2DKaufman%2C+Tal+Haklay%2C+Michael+Hanna%2C+Jing+Huang%2C+Rohan+Gupta%2C+Yaniv+Nikankin%2C+Ha)
202. Jord Nguyen, Khiem Hoang, Carlo Leonardo Attubato, and Felix Hofstätter. 2025b. Probing and steering evaluation awareness of language models, july 2025. URL http://arxiv.org/abs/2507.01786. — [arXiv:2507.01786](https://arxiv.org/abs/2507.01786) [↗](http://arxiv.org/abs/2507.01786)
203. Trang Nguyen, Jackson Michaels, Madalina Fiterau, and David Jensen. 2025c. Challenges in understanding modality conflict in vision-language models. arXiv preprint arXiv:2509.02805. — [arXiv:2509.02805](https://arxiv.org/abs/2509.02805) [find](https://scholar.google.com/scholar?q=Trang+Nguyen%2C+Jackson+Michaels%2C+Madalina+Fiterau%2C+and+David+Jensen%2E+2025c%2E+Challenges+in+understanding+modality+conflict+in+vision%2Dlanguage+models%2E+arXiv+preprint+arXiv%3A2509%2E02805%2E+%E2%80%94+%5BarXiv%3A2509%2E02805)
204. Tuc Nguyen and Thai Le. 2026. Atlas: Adaptive test-time latent steering with external verifiers for enhancing llms reasoning. arXiv preprint arXiv:2601.03093. — [arXiv:2601.03093](https://arxiv.org/abs/2601.03093) [find](https://scholar.google.com/scholar?q=Tuc+Nguyen+and+Thai+Le%2E+2026%2E+Atlas%3A+Adaptive+test%2Dtime+latent+steering+with+external+verifiers+for+enhancing+llms+reasoning%2E+arXiv+preprint+arXiv%3A2601%2E03093%2E+%E2%80%94+%5BarXiv%3A2601%2E03093%5D%28https%3A%2F%2Farxiv%2Eorg%2Fab)
205. Ercong Nie, Helmut Schmid, and Hinrich Schuetze. 2025. Mechanistic understanding and mitigation of language confusion in English-centric large language models. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 690–706, Suzhou, China. Association for Computational Lingui… [find](https://scholar.google.com/scholar?q=Ercong+Nie%2C+Helmut+Schmid%2C+and+Hinrich+Schuetze%2E+2025%2E+Mechanistic+understanding+and+mitigation+of+language+confusion+in+English%2Dcentric+large+language+models%2E+In+Findings+of+the+Association+for+Compu)
206. Aashiq Muhamed, Jacopo Bonato, Mona T. Diab, and Virginia Smith. 2025a. Saes Can improve unlearning: Dynamic sparse autoencoder guardrails for precision unlearning in llms. CoRR, abs/2504.08192. — [arXiv:2504.08192](https://arxiv.org/abs/2504.08192) [find](https://scholar.google.com/scholar?q=Aashiq+Muhamed%2C+Jacopo+Bonato%2C+Mona+T%2E+Diab%2C+and+Virginia+Smith%2E+2025a%2E+Saes+Can+improve+unlearning%3A+Dynamic+sparse+autoencoder+guardrails+for+precision+unlearning+in+llms%2E+CoRR%2C+abs%2F2504%2E08192%2E+%E2%80%94+%5Bar)
207. Yaniv Nikankin, Dana Arad, Yossi Gandelsman, and Yonatan Belinkov. 2025a. Same task, different circuits: Disentangling modality-specific mechanisms in vlms. arXiv preprint arXiv:2506.09047. — [arXiv:2506.09047](https://arxiv.org/abs/2506.09047) [find](https://scholar.google.com/scholar?q=Yaniv+Nikankin%2C+Dana+Arad%2C+Yossi+Gandelsman%2C+and+Yonatan+Belinkov%2E+2025a%2E+Same+task%2C+different+circuits%3A+Disentangling+modality%2Dspecific+mechanisms+in+vlms%2E+arXiv+preprint+arXiv%3A2506%2E09047%2E+%E2%80%94+%5BarXiv%3A2)
208. Aashiq Muhamed, Mona Diab, and Virginia Smith. 2025b. Decoding dark matter: Specialized sparse autoencoders for interpreting rare concepts in foundation models. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 1604– 1635. [find](https://scholar.google.com/scholar?q=Aashiq+Muhamed%2C+Mona+Diab%2C+and+Virginia+Smith%2E+2025b%2E+Decoding+dark+matter%3A+Specialized+sparse+autoencoders+for+interpreting+rare+concepts+in+foundation+models%2E+In+Findings+of+the+Association+for+Comp)
209. Yaniv Nikankin, Anja Reusch, Aaron Mueller, and Yonatan Belinkov. 2025b. [Arithmetic without algorithms: Language models solve math with a bag of heuristics](https://md.research.elamur.ai/free/en/papers/nikankin-2024-arithmetic-algorithms-language-models-solve/). In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net.
210. Aashiq Muhamed and Virginia Smith. 2025. The geometry of forgetting: Analyzing machine unlearning through local learning coefficients. In ICML 2025 Workshop on Reliable and Responsible Foundation Models. [find](https://scholar.google.com/scholar?q=Aashiq+Muhamed+and+Virginia+Smith%2E+2025%2E+The+geometry+of+forgetting%3A+Analyzing+machine+unlearning+through+local+learning+coefficients%2E+In+ICML+2025+Workshop+on+Reliable+and+Responsible+Foundation+Mode)
211. Jingcheng Niu, Xingdi Yuan, Tong Wang, Hamidreza Saghir, and Amir H. Abdi. 2025. Llama see, llama do: A mechanistic perspective on contextual entrainment and distraction in LLMs. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages… [find](https://scholar.google.com/scholar?q=Jingcheng+Niu%2C+Xingdi+Yuan%2C+Tong+Wang%2C+Hamidreza+Saghir%2C+and+Amir+H%2E+Abdi%2E+2025%2E+Llama+see%2C+llama+do%3A+A+mechanistic+perspective+on+contextual+entrainment+and+distraction+in+LLMs%2E+In+Proceedings+of+the)
212. Neel Nanda. 2023. Attribution patching: Activation patching at industrial scale. URL: [https://www.neel-nanda](https://www.neel-nanda). io/mechanistic-interpretability/attribution-patching. nostalgebraist. 2020. Interpreting GPT: the logit lens. [↗](https://www.neel-nanda)
213. Neel Nanda, Lawrence Chan, Tom Lieberum, Jess Smith, and Jacob Steinhardt. 2023. [Progress measures for grokking via mechanistic interpretability](https://md.research.elamur.ai/free/en/papers/nanda-2023-progress-measures-grokking-mechanistic-interpretability/). In The Eleventh International Conference on Learning Representations.
214. Pascal Notsawo Jr, Hattie Zhou, Mohammad Pezeshki, Irina Rish, Guillaume Dumas, and 1 others. 2023. Predicting grokking long before it happens: A look into the loss landscape of models which grok. arXiv preprint arXiv:2306.13253. — [arXiv:2306.13253](https://arxiv.org/abs/2306.13253) [find](https://scholar.google.com/scholar?q=Pascal+Notsawo+Jr%2C+Hattie+Zhou%2C+Mohammad+Pezeshki%2C+Irina+Rish%2C+Guillaume+Dumas%2C+and+1+others%2E+2023%2E+Predicting+grokking+long+before+it+happens%3A+A+look+into+the+loss+landscape+of+models+which+grok%2E+arX)
215. Matthew Lyle Olson, Neale Ratzlaff, Musashi Hinck, Man Luo, Sungduk Yu, Chendi Xue, and Vasudev Lal. 2025. Probing semantic routing in large mixture-of-expert models. In Findings of the Association for Computational Linguistics: EMNLP 2025, Suzhou, China, November 4-9, 2025, pages 18263–18278. Assoc… [find](https://scholar.google.com/scholar?q=Matthew+Lyle+Olson%2C+Neale+Ratzlaff%2C+Musashi+Hinck%2C+Man+Luo%2C+Sungduk+Yu%2C+Chendi+Xue%2C+and+Vasudev+Lal%2E+2025%2E+Probing+semantic+routing+in+large+mixture%2Dof%2Dexpert+models%2E+In+Findings+of+the+Association+fo)
216. Catherine Olsson, Nelson Elhage, Neel Nanda, Nicholas Joseph, Nova DasSarma, Tom Henighan, Ben Mann, Amanda Askell, Yuntao Bai, Anna Chen, Tom Conerly, Dawn Drain, Deep Ganguli, Zac Hatfield-Dodds, Danny Hernandez, Scott Johnston, Andy Jones, Jackson Kernion, Liane Lovitt, and 7 others. 2022. [In-con…](https://md.research.elamur.ai/free/en/papers/olsson-2022-context-learning-induction-heads/)
217. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, and 262 others.… — [GPT-4 Technical Report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/)
218. Hadas Orgad, Michael Toker, Zorik Gekhman, Roi Reichart, Idan Szpektor, Hadas Kotek, and Yonatan Belinkov. 2025. [LLMs know more than they show: On the intrinsic representation of LLM hallucinations](https://md.research.elamur.ai/free/en/papers/orgad-2024-llms-know-more-than-they/). In The Thirteenth International Conference on Learning Representations.
219. Mateusz Pach, Shyamgopal Karthik, Quentin Bouniot, Serge Belongie, and Zeynep Akata. 2025. Sparse autoencoders learn monosemantic features in vision-language models. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Mateusz+Pach%2C+Shyamgopal+Karthik%2C+Quentin+Bouniot%2C+Serge+Belongie%2C+and+Zeynep+Akata%2E+2025%2E+Sparse+autoencoders+learn+monosemantic+features+in+vision%2Dlanguage+models%2E+In+The+Thirty%2Dninth+Annual+Confere)
220. Tsung-Min Pai, Jui-I Wang, Li-Chun Lu, Shao-Hua Sun, Hung-Yi Lee, and Kai-Wei Chang. 2025. [Billy: Steering large language models via merging persona vectors for creative generation](https://md.research.elamur.ai/free/en/papers/pai-2026-billy-steering-large-language-models/). arXiv preprint arXiv:2510.10157.
221. Koyena Pal, Jiuding Sun, Andrew Yuan, Byron C Wallace, and David Bau. 2023. [Future lens: Anticipating subsequent tokens from a single hidden state](https://md.research.elamur.ai/free/en/papers/pal-2023-future-lens-anticipating-subsequent-tokens/). In Proceedings of the 27th Conference on Computational Natural Language Learning (CoNLL), pages 548– 560.
222. Wenbo Pan, Zhichao Liu, Qiguang Chen, Xiangyang Zhou, Yu Haining, and Xiaohua Jia. 2025. The hidden dimensions of LLM alignment: A multidimensional analysis of orthogonal safety directions. In Forty-second International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Wenbo+Pan%2C+Zhichao+Liu%2C+Qiguang+Chen%2C+Xiangyang+Zhou%2C+Yu+Haining%2C+and+Xiaohua+Jia%2E+2025%2E+The+hidden+dimensions+of+LLM+alignment%3A+A+multidimensional+analysis+of+orthogonal+safety+directions%2E+In+Forty%2Ds)
223. Abhishek Panigrahi, Nikunj Saunshi, Haoyu Zhao, and Sanjeev Arora. 2023. Task-specific skill localization in fine-tuned language models. Preprint, ICML:2302.06600. [find](https://scholar.google.com/scholar?q=Abhishek+Panigrahi%2C+Nikunj+Saunshi%2C+Haoyu+Zhao%2C+and+Sanjeev+Arora%2E+2023%2E+Task%2Dspecific+skill+localization+in+fine%2Dtuned+language+models%2E+Preprint%2C+ICML%3A2302%2E06600%2E)
224. Avi Parrack, Carlo Leonardo Attubato, and Stefan Heimersheim. 2025. Benchmarking deception probes via black-to-white performance boosts. arXiv preprint arXiv:2507.12691. — [arXiv:2507.12691](https://arxiv.org/abs/2507.12691) [find](https://scholar.google.com/scholar?q=Avi+Parrack%2C+Carlo+Leonardo+Attubato%2C+and+Stefan+Heimersheim%2E+2025%2E+Benchmarking+deception+probes+via+black%2Dto%2Dwhite+performance+boosts%2E+arXiv+preprint+arXiv%3A2507%2E12691%2E+%E2%80%94+%5BarXiv%3A2507%2E12691%5D%28https%3A%2F%2Fa)
225. Fred Philippy, Siwen Guo, and Shohreh Haddadan. 2023. Identifying the correlation between language distance and cross-lingual transfer in a multilingual representation space. In Proceedings of the 5th Workshop on Research in Computational Linguistic Typology and Multilingual NLP, pages 22–29, Dubrov… [find](https://scholar.google.com/scholar?q=Fred+Philippy%2C+Siwen+Guo%2C+and+Shohreh+Haddadan%2E+2023%2E+Identifying+the+correlation+between+language+distance+and+cross%2Dlingual+transfer+in+a+multilingual+representation+space%2E+In+Proceedings+of+the+5th)
226. Anirudh Phukan, Divyansh, Harshit Kumar Morj, Vaishnavi, Apoorv Saxena, and Koustava Goswami. 2025. Beyond logit lens: Contextual embeddings for robust hallucination detection & grounding in VLMs. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Com… [find](https://scholar.google.com/scholar?q=Anirudh+Phukan%2C+Divyansh%2C+Harshit+Kumar+Morj%2C+Vaishnavi%2C+Apoorv+Saxena%2C+and+Koustava+Goswami%2E+2025%2E+Beyond+logit+lens%3A+Contextual+embeddings+for+robust+hallucination+detection+%26+grounding+in+VLMs%2E+In+)
227. Joris Postmus and Steven Abreu. 2024. Steering large language models using conceptors: Improving addition-based activation engineering. arXiv preprint arXiv:2410.16314. — [arXiv:2410.16314](https://arxiv.org/abs/2410.16314) [find](https://scholar.google.com/scholar?q=Joris+Postmus+and+Steven+Abreu%2E+2024%2E+Steering+large+language+models+using+conceptors%3A+Improving+addition%2Dbased+activation+engineering%2E+arXiv+preprint+arXiv%3A2410%2E16314%2E+%E2%80%94+%5BarXiv%3A2410%2E16314%5D%28https%3A%2F%2Far)
228. Daniele Potertì, Andrea Seveso, and Fabio Mercorio. 2025. Can role vectors affect LLM behaviour? In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 17735–17747, Suzhou, China. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Daniele+Potert%C3%AC%2C+Andrea+Seveso%2C+and+Fabio+Mercorio%2E+2025%2E+Can+role+vectors+affect+LLM+behaviour%3F+In+Findings+of+the+Association+for+Computational+Linguistics%3A+EMNLP+2025%2C+pages+17735%E2%80%9317747%2C+Suzhou%2C+Ch)
229. Nikhil Prakash, Tamar Rott Shaham, Tal Haklay, Yonatan Belinkov, and David Bau. 2024. Fine-tuning enhances existing mechanisms: A case study on entity tracking. arXiv preprint arXiv:2402.14811. — [arXiv:2402.14811](https://arxiv.org/abs/2402.14811) [find](https://scholar.google.com/scholar?q=Nikhil+Prakash%2C+Tamar+Rott+Shaham%2C+Tal+Haklay%2C+Yonatan+Belinkov%2C+and+David+Bau%2E+2024%2E+Fine%2Dtuning+enhances+existing+mechanisms%3A+A+case+study+on+entity+tracking%2E+arXiv+preprint+arXiv%3A2402%2E14811%2E+%E2%80%94+%5BarX)
230. Hu Qiye, Zhou Hao, and Yu RuoXi. 2024. Exploring grokking: Experimental and mechanistic investigations. arXiv preprint arXiv:2412.10898. — [arXiv:2412.10898](https://arxiv.org/abs/2412.10898) [find](https://scholar.google.com/scholar?q=Hu+Qiye%2C+Zhou+Hao%2C+and+Yu+RuoXi%2E+2024%2E+Exploring+grokking%3A+Experimental+and+mechanistic+investigations%2E+arXiv+preprint+arXiv%3A2412%2E10898%2E+%E2%80%94+%5BarXiv%3A2412%2E10898%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2412%2E10898%29)
231. Philip Quirke and Fazl Barez. 2024. Understanding addition in transformers. In The Twelfth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Philip+Quirke+and+Fazl+Barez%2E+2024%2E+Understanding+addition+in+transformers%2E+In+The+Twelfth+International+Conference+on+Learning+Representations%2E)
232. Qwen, :, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, and 25 others. 2025. Qwen2.5 technical report. Preprint, arXiv:2412.15115. [find](https://scholar.google.com/scholar?q=Qwen%2C+%3A%2C+An+Yang%2C+Baosong+Yang%2C+Beichen+Zhang%2C+Binyuan+Hui%2C+Bo+Zheng%2C+Bowen+Yu%2C+Chengyuan+Li%2C+Dayiheng+Liu%2C+Fei+Huang%2C+Haoran+Wei%2C+Huan+Lin%2C+Jian+Yang%2C+Jianhong+Tu%2C+Jianwei+Zhang%2C+Jianxin+Yang%2C+Jiaxi+)
233. Daking Rai, Yilun Zhou, Shi Feng, Abulhair Saparov, and Ziyu Yao. 2024. A practical review of mechanistic interpretability for transformer-based language models. arXiv preprint arXiv:2407.02646. [find](https://scholar.google.com/scholar?q=Daking+Rai%2C+Yilun+Zhou%2C+Shi+Feng%2C+Abulhair+Saparov%2C+and+Ziyu+Yao%2E+2024%2E+A+practical+review+of+mechanistic+interpretability+for+transformer%2Dbased+language+models%2E+arXiv+preprint+arXiv%3A2407%2E02646%2E)
234. Bianca Raimondi, Daniela Dalbagno, and Maurizio Gabbrielli. 2025. Analysing Moral Bias in Finetuned LLMs through Mechanistic Interpretability. Preprint, arXiv:2510.12229. — [arXiv:2510.12229](https://arxiv.org/abs/2510.12229) [find](https://scholar.google.com/scholar?q=Bianca+Raimondi%2C+Daniela+Dalbagno%2C+and+Maurizio+Gabbrielli%2E+2025%2E+Analysing+Moral+Bias+in+Finetuned+LLMs+through+Mechanistic+Interpretability%2E+Preprint%2C+arXiv%3A2510%2E12229%2E+%E2%80%94+%5BarXiv%3A2510%2E12229%5D%28https%3A%2F%2F)
235. Senthooran Rajamanoharan, Arthur Conmy, Lewis Smith, Tom Lieberum, Vikrant Varma, János Kramár, Rohin Shah, and Neel Nanda. 2024a. [Improving dictionary learning with gated sparse autoencoders](https://md.research.elamur.ai/free/en/papers/rajamanoharan-2024-improving-dictionary-learning-gated-sparse/). arXiv preprint arXiv:2404.16014.
236. Chenming Shang, Shiji Zhou, Hengyuan Zhang, Xinzhe Ni, Yujiu Yang, and Yuwang Wang. 2024b. Incremental residual concept bottleneck models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11030–11040. [find](https://scholar.google.com/scholar?q=Chenming+Shang%2C+Shiji+Zhou%2C+Hengyuan+Zhang%2C+Xinzhe+Ni%2C+Yujiu+Yang%2C+and+Yuwang+Wang%2E+2024b%2E+Incremental+residual+concept+bottleneck+models%2E+In+Proceedings+of+the+IEEE%2FCVF+Conference+on+Computer+Vision+)
237. Senthooran Rajamanoharan, Tom Lieberum, Nicolas Sonnerat, Arthur Conmy, Vikrant Varma, János Kramár, and Neel Nanda. 2024b. Jumping ahead: Improving reconstruction fidelity with jumprelu sparse autoencoders. arXiv preprint arXiv:2407.14435. — [arXiv:2407.14435](https://arxiv.org/abs/2407.14435) [find](https://scholar.google.com/scholar?q=Senthooran+Rajamanoharan%2C+Tom+Lieberum%2C+Nicolas+Sonnerat%2C+Arthur+Conmy%2C+Vikrant+Varma%2C+J%C3%A1nos+Kram%C3%A1r%2C+and+Neel+Nanda%2E+2024b%2E+Jumping+ahead%3A+Improving+reconstruction+fidelity+with+jumprelu+sparse+autoen)
238. Jiandong Shao, Yao Lu, and Jianfei Yang. 2025. Benford’s curse: Tracing digit bias to numerical hallucination in llms. arXiv preprint arXiv:2506.01734. — [arXiv:2506.01734](https://arxiv.org/abs/2506.01734) [find](https://scholar.google.com/scholar?q=Jiandong+Shao%2C+Yao+Lu%2C+and+Jianfei+Yang%2E+2025%2E+Benford%E2%80%99s+curse%3A+Tracing+digit+bias+to+numerical+hallucination+in+llms%2E+arXiv+preprint+arXiv%3A2506%2E01734%2E+%E2%80%94+%5BarXiv%3A2506%2E01734%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2506%2E)
239. Navin Ranjan and Andreas Savakis. 2025. Mix-qvit: Mixed-precision vision transformer quantization driven by layer importance and quantization sensitivity. arXiv preprint arXiv:2501.06357. — [arXiv:2501.06357](https://arxiv.org/abs/2501.06357) [find](https://scholar.google.com/scholar?q=Navin+Ranjan+and+Andreas+Savakis%2E+2025%2E+Mix%2Dqvit%3A+Mixed%2Dprecision+vision+transformer+quantization+driven+by+layer+importance+and+quantization+sensitivity%2E+arXiv+preprint+arXiv%3A2501%2E06357%2E+%E2%80%94+%5BarXiv%3A250)
240. Tilman Räuker, Anson Ho, Stephen Casper, and Dylan Hadfield-Menell. 2023. Toward transparent ai: A survey on interpreting the inner structures of deep neural networks. In 2023 ieee conference on secure and trustworthy machine learning (satml), pages 464– 483. IEEE. [find](https://scholar.google.com/scholar?q=Tilman+R%C3%A4uker%2C+Anson+Ho%2C+Stephen+Casper%2C+and+Dylan+Hadfield%2DMenell%2E+2023%2E+Toward+transparent+ai%3A+A+survey+on+interpreting+the+inner+structures+of+deep+neural+networks%2E+In+2023+ieee+conference+on+secur)
241. Santhosh Kumar Ravindran. 2025. Adversarial activation patching: A framework for detecting and mitigating emergent deception in safety-aligned transformers. arXiv preprint arXiv:2507.09406. — [arXiv:2507.09406](https://arxiv.org/abs/2507.09406) [find](https://scholar.google.com/scholar?q=Santhosh+Kumar+Ravindran%2E+2025%2E+Adversarial+activation+patching%3A+A+framework+for+detecting+and+mitigating+emergent+deception+in+safety%2Daligned+transformers%2E+arXiv+preprint+arXiv%3A2507%2E09406%2E+%E2%80%94+%5BarXiv%3A2)
242. Yuqi Ren, Renren Jin, Tongxuan Zhang, and Deyi Xiong. 2025. [Do large language models mirror cognitive language processing?](https://md.research.elamur.ai/free/en/papers/ren-2025-do-large-language-models-mirror/) In Proceedings of the 31st International Conference on Computational Linguistics, pages 2988–3001.
243. Nina Rimsky, Nick Gabrieli, Julian Schulz, Meg Tong, Evan Hubinger, and Alexander Turner. 2024. Steering llama 2 via contrastive activation addition. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15504–15522, Bangkok, Thaila… [find](https://scholar.google.com/scholar?q=Nina+Rimsky%2C+Nick+Gabrieli%2C+Julian+Schulz%2C+Meg+Tong%2C+Evan+Hubinger%2C+and+Alexander+Turner%2E+2024%2E+Steering+llama+2+via+contrastive+activation+addition%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+A)
244. Mansi Sakarvadia, Arham Khan, Aswathy Ajith, Daniel Grzenda, Nathaniel Hudson, André Bauer, Kyle Chard, and Ian Foster. 2023. Attention lens: A tool for mechanistically interpreting the attention head information retrieval mechanism. arXiv preprint arXiv:2310.16270. — [arXiv:2310.16270](https://arxiv.org/abs/2310.16270) [find](https://scholar.google.com/scholar?q=Mansi+Sakarvadia%2C+Arham+Khan%2C+Aswathy+Ajith%2C+Daniel+Grzenda%2C+Nathaniel+Hudson%2C+Andr%C3%A9+Bauer%2C+Kyle+Chard%2C+and+Ian+Foster%2E+2023%2E+Attention+lens%3A+A+tool+for+mechanistically+interpreting+the+attention+head)
245. Alexander Sergeev and Evgeny Kotelnikov. 2025. Optimizing multimodal language models through attention-based interpretability. Preprint, arXiv:2511.23375. — [arXiv:2511.23375](https://arxiv.org/abs/2511.23375) [find](https://scholar.google.com/scholar?q=Alexander+Sergeev+and+Evgeny+Kotelnikov%2E+2025%2E+Optimizing+multimodal+language+models+through+attention%2Dbased+interpretability%2E+Preprint%2C+arXiv%3A2511%2E23375%2E+%E2%80%94+%5BarXiv%3A2511%2E23375%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F25)
246. Adam Shai, Lucas Teixeira, Alexander Oldenziel, Sarah Marzen, and Paul Riechers. 2024. Transformers represent belief state geometry in their residual stream. Advances in Neural Information Processing Systems, 37:75012–75034. [find](https://scholar.google.com/scholar?q=Adam+Shai%2C+Lucas+Teixeira%2C+Alexander+Oldenziel%2C+Sarah+Marzen%2C+and+Paul+Riechers%2E+2024%2E+Transformers+represent+belief+state+geometry+in+their+residual+stream%2E+Advances+in+Neural+Information+Processing+)
247. Chenming Shang, Hengyuan Zhang, Hao Wen, and Yujiu Yang. 2024a. Understanding multimodal deep neural networks: A concept selection view. arXiv preprint arXiv:2404.08964. — [arXiv:2404.08964](https://arxiv.org/abs/2404.08964) [find](https://scholar.google.com/scholar?q=Chenming+Shang%2C+Hengyuan+Zhang%2C+Hao+Wen%2C+and+Yujiu+Yang%2E+2024a%2E+Understanding+multimodal+deep+neural+networks%3A+A+concept+selection+view%2E+arXiv+preprint+arXiv%3A2404%2E08964%2E+%E2%80%94+%5BarXiv%3A2404%2E08964%5D%28https%3A%2F%2Fa)
248. Vansh Sharma and Venkat Raman. 2025. Steering conceptual bias via transformer latent-subspace activation. Preprint, arXiv:2506.18887. — [arXiv:2506.18887](https://arxiv.org/abs/2506.18887) [find](https://scholar.google.com/scholar?q=Vansh+Sharma+and+Venkat+Raman%2E+2025%2E+Steering+conceptual+bias+via+transformer+latent%2Dsubspace+activation%2E+Preprint%2C+arXiv%3A2506%2E18887%2E+%E2%80%94+%5BarXiv%3A2506%2E18887%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2506%2E18887%29)
249. Noam Shazeer, Azalia Mirhoseini, Krzysztof Maziarz, Andy Davis, Quoc V. Le, Geoffrey E. Hinton, and Jeff Dean. 2017. Outrageously large neural networks: The sparsely-gated mixture-of-experts layer. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2… [find](https://scholar.google.com/scholar?q=Noam+Shazeer%2C+Azalia+Mirhoseini%2C+Krzysztof+Maziarz%2C+Andy+Davis%2C+Quoc+V%2E+Le%2C+Geoffrey+E%2E+Hinton%2C+and+Jeff+Dean%2E+2017%2E+Outrageously+large+neural+networks%3A+The+sparsely%2Dgated+mixture%2Dof%2Dexperts+layer%2E+In)
250. Anushka Shelke, Riya Savant, and Raviraj Joshi. 2024. Towards building efficient sentence bert models using layer pruning. In Proceedings of the 38th Pacific Asia Conference on Language, Information and Computation, pages 720–725. [find](https://scholar.google.com/scholar?q=Anushka+Shelke%2C+Riya+Savant%2C+and+Raviraj+Joshi%2E+2024%2E+Towards+building+efficient+sentence+bert+models+using+layer+pruning%2E+In+Proceedings+of+the+38th+Pacific+Asia+Conference+on+Language%2C+Information+a)
251. Dan Shi, Renren Jin, Tianhao Shen, Weilong Dong, Xinwei Wu, and Deyi Xiong. 2024. IRCAN: mitigating knowledge conflicts in LLM generation via identifying and reweighting context-aware neurons. In Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing… [find](https://scholar.google.com/scholar?q=Dan+Shi%2C+Renren+Jin%2C+Tianhao+Shen%2C+Weilong+Dong%2C+Xinwei+Wu%2C+and+Deyi+Xiong%2E+2024%2E+IRCAN%3A+mitigating+knowledge+conflicts+in+LLM+generation+via+identifying+and+reweighting+context%2Daware+neurons%2E+In+Adva)
252. Avanti Shrikumar, Peyton Greenside, and Anshul Kundaje. 2017. [Learning important features through propagating activation differences](https://md.research.elamur.ai/free/en/papers/shrikumar-2017-learning-important-features-through-propagating/). In Proceedings of the 34th International Conference on Machine Learning (ICML), volume 70 of Proceedings of Machine Learning Research.
253. Dong Shu, Xuansheng Wu, Haiyan Zhao, Daking Rai, Ziyu Yao, Ninghao Liu, and Mengnan Du. 2025. A survey on sparse autoencoders: Interpreting the internal mechanisms of large language models. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 1690–1712, Suzhou, China. Asso… [find](https://scholar.google.com/scholar?q=Dong+Shu%2C+Xuansheng+Wu%2C+Haiyan+Zhao%2C+Daking+Rai%2C+Ziyu+Yao%2C+Ninghao+Liu%2C+and+Mengnan+Du%2E+2025%2E+A+survey+on+sparse+autoencoders%3A+Interpreting+the+internal+mechanisms+of+large+language+models%2E+In+Finding)
254. Aaditya K Singh, Ted Moskovitz, Felix Hill, Stephanie CY Chan, and Andrew M Saxe. 2024. What needs to go right for an induction head? a mechanistic study of in-context learning circuits and their formation. In Proceedings of the 41st International Conference on Machine Learning, pages 45637–45662. [find](https://scholar.google.com/scholar?q=Aaditya+K+Singh%2C+Ted+Moskovitz%2C+Felix+Hill%2C+Stephanie+CY+Chan%2C+and+Andrew+M+Saxe%2E+2024%2E+What+needs+to+go+right+for+an+induction+head%3F+a+mechanistic+study+of+in%2Dcontext+learning+circuits+and+their+form)
255. Viacheslav Sinii, Alexey Gorbatovski, Artem Cherepanov, Boris Shaposhnikov, Nikita Balagansky, and Daniil Gavrilov. 2025. Steering llm reasoning through bias-only adaptation. Preprint, arXiv:2505.18706. [find](https://scholar.google.com/scholar?q=Viacheslav+Sinii%2C+Alexey+Gorbatovski%2C+Artem+Cherepanov%2C+Boris+Shaposhnikov%2C+Nikita+Balagansky%2C+and+Daniil+Gavrilov%2E+2025%2E+Steering+llm+reasoning+through+bias%2Donly+adaptation%2E+Preprint%2C+arXiv%3A2505%2E1870)
256. Daniel Smilkov, Nikhil Thorat, Been Kim, Fernanda B. Viégas, and Martin Wattenberg. 2017. Smoothgrad: removing noise by adding noise. CoRR, abs/1706.03825. — [arXiv:1706.03825](https://arxiv.org/abs/1706.03825) [find](https://scholar.google.com/scholar?q=Daniel+Smilkov%2C+Nikhil+Thorat%2C+Been+Kim%2C+Fernanda+B%2E+Vi%C3%A9gas%2C+and+Martin+Wattenberg%2E+2017%2E+Smoothgrad%3A+removing+noise+by+adding+noise%2E+CoRR%2C+abs%2F1706%2E03825%2E+%E2%80%94+%5BarXiv%3A1706%2E03825%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1)
257. Zunhai Su, Qingyuan Li, Hao Zhang, YuLei Qian, Yuchen Xie, and Kehong Yuan. 2025d. Unveiling super experts in mixture-of-experts large language models. arXiv preprint arXiv:2507.23279. — [arXiv:2507.23279](https://arxiv.org/abs/2507.23279) [find](https://scholar.google.com/scholar?q=Zunhai+Su%2C+Qingyuan+Li%2C+Hao+Zhang%2C+YuLei+Qian%2C+Yuchen+Xie%2C+and+Kehong+Yuan%2E+2025d%2E+Unveiling+super+experts+in+mixture%2Dof%2Dexperts+large+language+models%2E+arXiv+preprint+arXiv%3A2507%2E23279%2E+%E2%80%94+%5BarXiv%3A2507%2E2)
258. Weixi Song, Zuchao Li, Lefei Zhang, Hai Zhao, and Bo Du. 2024. Sparse is enough in fine-tuning pre-trained large language models. In Proceedings of the 41st International Conference on Machine Learning, ICML’24. JMLR.org. [find](https://scholar.google.com/scholar?q=Weixi+Song%2C+Zuchao+Li%2C+Lefei+Zhang%2C+Hai+Zhao%2C+and+Bo+Du%2E+2024%2E+Sparse+is+enough+in+fine%2Dtuning+pre%2Dtrained+large+language+models%2E+In+Proceedings+of+the+41st+International+Conference+on+Machine+Learnin)
259. Zunhai Su and Kehong Yuan. 2025. Kvsink: Understanding and enhancing the preservation of attention sinks in kv cache quantization for llms. arXiv preprint arXiv:2508.04257. — [arXiv:2508.04257](https://arxiv.org/abs/2508.04257) [find](https://scholar.google.com/scholar?q=Zunhai+Su+and+Kehong+Yuan%2E+2025%2E+Kvsink%3A+Understanding+and+enhancing+the+preservation+of+attention+sinks+in+kv+cache+quantization+for+llms%2E+arXiv+preprint+arXiv%3A2508%2E04257%2E+%E2%80%94+%5BarXiv%3A2508%2E04257%5D%28https%3A)
260. Samuel Soo, Chen Guang, Wesley Teng, Chandrasekaran Balaganesh, Tan Guoxian, and Yan Ming. 2025. Interpretable steering of large language models with feature guided activation additions. arXiv preprint arXiv:2501.09929. — [arXiv:2501.09929](https://arxiv.org/abs/2501.09929) [find](https://scholar.google.com/scholar?q=Samuel+Soo%2C+Chen+Guang%2C+Wesley+Teng%2C+Chandrasekaran+Balaganesh%2C+Tan+Guoxian%2C+and+Yan+Ming%2E+2025%2E+Interpretable+steering+of+large+language+models+with+feature+guided+activation+additions%2E+arXiv+preprin)
261. Xavier Suau, Pieter Delobelle, Katherine Metcalf, Armand Joulin, Nicholas Apostoloff, Luca Zappella, and Pau Rodriguez. 2024. Whispering experts: Neural interventions for toxicity mitigation in language models. In Forty-first International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Xavier+Suau%2C+Pieter+Delobelle%2C+Katherine+Metcalf%2C+Armand+Joulin%2C+Nicholas+Apostoloff%2C+Luca+Zappella%2C+and+Pau+Rodriguez%2E+2024%2E+Whispering+experts%3A+Neural+interventions+for+toxicity+mitigation+in+langua)
262. Divyansh Srivastava, Ge Yan, and Lily Weng. 2024. Vlg-cbm: Training concept bottleneck models with vision-language guidance. Advances in Neural Information Processing Systems, 37:79057–79094. [find](https://scholar.google.com/scholar?q=Divyansh+Srivastava%2C+Ge+Yan%2C+and+Lily+Weng%2E+2024%2E+Vlg%2Dcbm%3A+Training+concept+bottleneck+models+with+vision%2Dlanguage+guidance%2E+Advances+in+Neural+Information+Processing+Systems%2C+37%3A79057%E2%80%9379094%2E)
263. Niklas Stoehr, Kevin Du, Vésteinn Snæbjarnarson, Robert West, Ryan Cotterell, and Aaron Schein. 2024. Activation scaling for steering and interpreting language models. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 8189–8200, Miami, Florida, USA. Association for Comp… [find](https://scholar.google.com/scholar?q=Niklas+Stoehr%2C+Kevin+Du%2C+V%C3%A9steinn+Sn%C3%A6bjarnarson%2C+Robert+West%2C+Ryan+Cotterell%2C+and+Aaron+Schein%2E+2024%2E+Activation+scaling+for+steering+and+interpreting+language+models%2E+In+Findings+of+the+Association+f)
264. Alessandro Stolfo, Vidhisha Balachandran, Safoora Yousefi, Eric Horvitz, and Besmira Nushi. 2025. Improving instruction-following in language models through activation steering. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Alessandro+Stolfo%2C+Vidhisha+Balachandran%2C+Safoora+Yousefi%2C+Eric+Horvitz%2C+and+Besmira+Nushi%2E+2025%2E+Improving+instruction%2Dfollowing+in+language+models+through+activation+steering%2E+In+The+Thirteenth+Inte)
265. Alessandro Stolfo, Yonatan Belinkov, and Mrinmaya Sachan. 2023. A mechanistic interpretation of arithmetic reasoning in language models using causal mediation analysis. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 7035–7052, Singapore. Association… [find](https://scholar.google.com/scholar?q=Alessandro+Stolfo%2C+Yonatan+Belinkov%2C+and+Mrinmaya+Sachan%2E+2023%2E+A+mechanistic+interpretation+of+arithmetic+reasoning+in+language+models+using+causal+mediation+analysis%2E+In+Proceedings+of+the+2023+Conf)
266. Jingran Su, Jingfan Chen, Hongxin Li, Yuntao Chen, Li Qing, and Zhaoxiang Zhang. 2025a. [Activation steering decoding: Mitigating hallucination in large vision-language models through bidirectional hidden state intervention](https://md.research.elamur.ai/free/en/papers/su-2025-activation-steering-decoding-mitigating-hallucination/). In Proceedings of the 63rd Annual Meeting of the Association for Computation…
267. Yi Su, Jiayi Zhang, Shu Yang, Xinhai Wang, Lijie Hu, and Di Wang. 2025b. Understanding how value neurons shape the generation of specified values in LLMs. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 9433–9452, Suzhou, China. Association for Computational Linguisti… [find](https://scholar.google.com/scholar?q=Yi+Su%2C+Jiayi+Zhang%2C+Shu+Yang%2C+Xinhai+Wang%2C+Lijie+Hu%2C+and+Di+Wang%2E+2025b%2E+Understanding+how+value+neurons+shape+the+generation+of+specified+values+in+LLMs%2E+In+Findings+of+the+Association+for+Computatio)
268. Zunhai Su, Zhe Chen, Wang Shen, Hanyu Wei, Linge Li, Huangqi Yu, and Kehong Yuan. 2025c. Rotatekv: Accurate and robust 2-bit kv cache quantization for llms via outlier-aware adaptive rotations. arXiv preprint arXiv:2501.16383. — [arXiv:2501.16383](https://arxiv.org/abs/2501.16383) [find](https://scholar.google.com/scholar?q=Zunhai+Su%2C+Zhe+Chen%2C+Wang+Shen%2C+Hanyu+Wei%2C+Linge+Li%2C+Huangqi+Yu%2C+and+Kehong+Yuan%2E+2025c%2E+Rotatekv%3A+Accurate+and+robust+2%2Dbit+kv+cache+quantization+for+llms+via+outlier%2Daware+adaptive+rotations%2E+arXiv+)
269. Chung-En Sun, Tuomas Oikarinen, Berk Ustun, and Tsui-Wei Weng. 2024a. Concept bottleneck large language models. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Chung%2DEn+Sun%2C+Tuomas+Oikarinen%2C+Berk+Ustun%2C+and+Tsui%2DWei+Weng%2E+2024a%2E+Concept+bottleneck+large+language+models%2E+In+The+Thirteenth+International+Conference+on+Learning+Representations%2E)
270. Mingjie Sun, Xinlei Chen, J Zico Kolter, and Zhuang Liu. 2024b. [Massive activations in large language models](https://md.research.elamur.ai/free/en/papers/sun-2024-massive-activations-large-language-models/). arXiv preprint arXiv:2402.17762.
271. Seungjong Sun, Seo Yeon Baek, and Jang Hyun Kim. 2025a. Personality vector: Modulating personality of large language models by model merging. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 24667– 24688, Suzhou, China. Association for Computational Li… [find](https://scholar.google.com/scholar?q=Seungjong+Sun%2C+Seo+Yeon+Baek%2C+and+Jang+Hyun+Kim%2E+2025a%2E+Personality+vector%3A+Modulating+personality+of+large+language+models+by+model+merging%2E+In+Proceedings+of+the+2025+Conference+on+Empirical+Methods)
272. Yucheng Sun, Alessandro Stolfo, and Mrinmaya Sachan. 2025b. Probing for arithmetic errors in language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 8122–8139, Suzhou, China. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Yucheng+Sun%2C+Alessandro+Stolfo%2C+and+Mrinmaya+Sachan%2E+2025b%2E+Probing+for+arithmetic+errors+in+language+models%2E+In+Proceedings+of+the+2025+Conference+on+Empirical+Methods+in+Natural+Language+Processing%2C)
273. Mukund Sundararajan, Ankur Taly, and Qiqi Yan. 2017. Axiomatic attribution for deep networks. In Proceedings of the 34th International Conference on Machine Learning, ICML 2017, Sydney, NSW, Australia, 6-11 August 2017, volume 70 of Proceedings of Machine Learning Research, pages 3319–3328. PMLR. [find](https://scholar.google.com/scholar?q=Mukund+Sundararajan%2C+Ankur+Taly%2C+and+Qiqi+Yan%2E+2017%2E+Axiomatic+attribution+for+deep+networks%2E+In+Proceedings+of+the+34th+International+Conference+on+Machine+Learning%2C+ICML+2017%2C+Sydney%2C+NSW%2C+Australia)
274. Aaquib Syed, Can Rager, and Arthur Conmy. 2024. [Attribution patching outperforms automated circuit discovery](https://md.research.elamur.ai/free/en/papers/syed-2023-attribution-patching-outperforms-automated-circuit/). In Proceedings of the 7th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP, pages 407–416, Miami, Florida, US. Association for Computational Linguistics.
275. Ala N. Tak, Amin Banayeeanzade, Anahita Bolourani, Mina Kian, Robin Jia, and Jonathan Gratch. 2025. Mechanistic interpretability of emotion inference in large language models. In Findings of the Association for Computational Linguistics: ACL 2025, pages 13090–13120, Vienna, Austria. Association for… [find](https://scholar.google.com/scholar?q=Ala+N%2E+Tak%2C+Amin+Banayeeanzade%2C+Anahita+Bolourani%2C+Mina+Kian%2C+Robin+Jia%2C+and+Jonathan+Gratch%2E+2025%2E+Mechanistic+interpretability+of+emotion+inference+in+large+language+models%2E+In+Findings+of+the+Assoc)
276. Ryota Takatsuki, Sonia Joseph, Ippei Fujisawa, and Ryota Kanai. 2025. Decoding vision transformers: the diffusion steering lens. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 4819–4824. [find](https://scholar.google.com/scholar?q=Ryota+Takatsuki%2C+Sonia+Joseph%2C+Ippei+Fujisawa%2C+and+Ryota+Kanai%2E+2025%2E+Decoding+vision+transformers%3A+the+diffusion+steering+lens%2E+In+Proceedings+of+the+Computer+Vision+and+Pattern+Recognition+Conferenc)
277. Shaomu Tan, Di Wu, and Christof Monz. 2024a. Neuron specialization: Leveraging intrinsic task modularity for multilingual machine translation. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 6506–6527, Miami, Florida, USA. Association for Computationa… [find](https://scholar.google.com/scholar?q=Shaomu+Tan%2C+Di+Wu%2C+and+Christof+Monz%2E+2024a%2E+Neuron+specialization%3A+Leveraging+intrinsic+task+modularity+for+multilingual+machine+translation%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+Method)
278. Yuqiao Tan, Minzheng Wang, Shizhu He, Huanxuan Liao, Chengfeng Zhao, Qiunan Lu, Tian Liang, Jun Zhao, and Kang Liu. 2025. Bottom-up policy optimization: Your language model policy secretly contains internal policies. Preprint, arXiv:2512.19673. — [arXiv:2512.19673](https://arxiv.org/abs/2512.19673) [find](https://scholar.google.com/scholar?q=Yuqiao+Tan%2C+Minzheng+Wang%2C+Shizhu+He%2C+Huanxuan+Liao%2C+Chengfeng+Zhao%2C+Qiunan+Lu%2C+Tian+Liang%2C+Jun+Zhao%2C+and+Kang+Liu%2E+2025%2E+Bottom%2Dup+policy+optimization%3A+Your+language+model+policy+secretly+contains+in)
279. Zhen Tan, Lu Cheng, Song Wang, Bo Yuan, Jundong Li, and Huan Liu. 2024b. Interpreting pretrained language models via concept bottlenecks. In Pacific-Asia Conference on Knowledge Discovery and Data Mining, pages 56–74. Springer. [find](https://scholar.google.com/scholar?q=Zhen+Tan%2C+Lu+Cheng%2C+Song+Wang%2C+Bo+Yuan%2C+Jundong+Li%2C+and+Huan+Liu%2E+2024b%2E+Interpreting+pretrained+language+models+via+concept+bottlenecks%2E+In+Pacific%2DAsia+Conference+on+Knowledge+Discovery+and+Data+Min)
280. Vimal Thilak, Etai Littwin, Shuangfei Zhai, Omid Saremi, Roni Paiss, and Joshua M Susskind. 2022. The slingshot mechanism: An empirical study of adaptive optimizers and the\\emph {Grokking Phenomenon}. In Has it Trained Yet? NeurIPS 2022 Workshop. [find](https://scholar.google.com/scholar?q=Vimal+Thilak%2C+Etai+Littwin%2C+Shuangfei+Zhai%2C+Omid+Saremi%2C+Roni+Paiss%2C+and+Joshua+M+Susskind%2E+2022%2E+The+slingshot+mechanism%3A+An+empirical+study+of+adaptive+optimizers+and+the%5C%5Cemph+%7BGrokking+Phenomenon%7D)
281. Dmitrii Troitskii, Koyena Pal, Chris Wendler, and Callum Stuart McDougall. 2025. Internal states before wait modulate reasoning patterns. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 18640–18649, Suzhou, China. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Dmitrii+Troitskii%2C+Koyena+Pal%2C+Chris+Wendler%2C+and+Callum+Stuart+McDougall%2E+2025%2E+Internal+states+before+wait+modulate+reasoning+patterns%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+)
282. Alexander Matt Turner, Lisa Thiergart, Gavin Leech, David Udell, Juan J. Vazquez, Ulisse Mini, and Monte MacDiarmid. 2024. [Steering language models with activation engineering](https://md.research.elamur.ai/free/en/papers/turner-2023-steering-language-models-activation-engineering/). Preprint, arXiv:2308.10248.
283. Florian Valade. 2024. Accelerating large language model inference with self-supervised early exits. Preprint, arXiv:2407.21082. — [arXiv:2407.21082](https://arxiv.org/abs/2407.21082) [find](https://scholar.google.com/scholar?q=Florian+Valade%2E+2024%2E+Accelerating+large+language+model+inference+with+self%2Dsupervised+early+exits%2E+Preprint%2C+arXiv%3A2407%2E21082%2E+%E2%80%94+%5BarXiv%3A2407%2E21082%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2407%2E21082%29)
284. Hanlin Tang, Yang Lin, Jing Lin, Qingsen Han, Shikuan Hong, Yiwu Yao, and Gongyi Wang. 2024a. Razorattention: Efficient kv cache compression through retrieval heads. arXiv preprint arXiv:2407.15891. — [arXiv:2407.15891](https://arxiv.org/abs/2407.15891) [find](https://scholar.google.com/scholar?q=Hanlin+Tang%2C+Yang+Lin%2C+Jing+Lin%2C+Qingsen+Han%2C+Shikuan+Hong%2C+Yiwu+Yao%2C+and+Gongyi+Wang%2E+2024a%2E+Razorattention%3A+Efficient+kv+cache+compression+through+retrieval+heads%2E+arXiv+preprint+arXiv%3A2407%2E15891%2E+%E2%80%94)
285. Teun van der Weij, Massimo Poesio, and Nandi Schoots. 2024. Extending activation steering to broad skills and multiple behaviours. arXiv preprint arXiv:2403.05767. — [arXiv:2403.05767](https://arxiv.org/abs/2403.05767) [find](https://scholar.google.com/scholar?q=Teun+van+der+Weij%2C+Massimo+Poesio%2C+and+Nandi+Schoots%2E+2024%2E+Extending+activation+steering+to+broad+skills+and+multiple+behaviours%2E+arXiv+preprint+arXiv%3A2403%2E05767%2E+%E2%80%94+%5BarXiv%3A2403%2E05767%5D%28https%3A%2F%2Farxiv%2Eo)
286. Tianyi Tang, Wenyang Luo, Haoyang Huang, Dongdong Zhang, Xiaolei Wang, Xin Zhao, Furu Wei, and Ji-Rong Wen. 2024b. [Language-specific neurons: The key to multilingual capabilities in large language models](https://md.research.elamur.ai/free/en/papers/tang-2024-language-specific-neurons-key-multilingual/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Vol…
287. Vikrant Varma, Rohin Shah, Zachary Kenton, János Kramár, and Ramana Kumar. 2023. Explaining grokking through circuit efficiency. arXiv preprint arXiv:2309.02390. — [arXiv:2309.02390](https://arxiv.org/abs/2309.02390) [find](https://scholar.google.com/scholar?q=Vikrant+Varma%2C+Rohin+Shah%2C+Zachary+Kenton%2C+J%C3%A1nos+Kram%C3%A1r%2C+and+Ramana+Kumar%2E+2023%2E+Explaining+grokking+through+circuit+efficiency%2E+arXiv+preprint+arXiv%3A2309%2E02390%2E+%E2%80%94+%5BarXiv%3A2309%2E02390%5D%28https%3A%2F%2Farxiv%2Eorg)
288. Xinyu Tang, Xiaolei Wang, Zhihao Lv, Yingqian Min, Wayne Xin Zhao, Binbin Hu, Ziqi Liu, and Zhiqiang Zhang. 2025. [Unlocking general long chain-of-thought reasoning capabilities of large language models via representation engineering](https://md.research.elamur.ai/free/en/papers/tang-2025-unlocking-general-long-chain-thought/). In Proceedings of the 63rd Annual Meeting of the Association for C…
289. Gemma Team, Thomas Mesnard, Cassidy Hardin, Robert Dadashi, Surya Bhupatiraju, Shreya Pathak, Laurent Sifre, Morgane Rivière, Mihir Sanjay Kale, Juliette Love, and 1 others. 2024. Gemma: Open models based on gemini research and technology. arXiv preprint arXiv:2403.08295. [find](https://scholar.google.com/scholar?q=Gemma+Team%2C+Thomas+Mesnard%2C+Cassidy+Hardin%2C+Robert+Dadashi%2C+Surya+Bhupatiraju%2C+Shreya+Pathak%2C+Laurent+Sifre%2C+Morgane+Rivi%C3%A8re%2C+Mihir+Sanjay+Kale%2C+Juliette+Love%2C+and+1+others%2E+2024%2E+Gemma%3A+Open+models+b)
290. Adly Templeton, Tom Conerly, Jonathan Marcus, Jack Lindsey, Trenton Bricken, Brian Chen, Adam Pearce, Craig Citro, Emmanuel Ameisen, Andy Jones, Hoagy Cunningham, Nicholas L Turner, Callum McDougall, Monte MacDiarmid, C. Daniel Freeman, Theodore R. Sumers, Edward Rees, Joshua Batson, Adam Jermyn, an… [find](https://scholar.google.com/scholar?q=Adly+Templeton%2C+Tom+Conerly%2C+Jonathan+Marcus%2C+Jack+Lindsey%2C+Trenton+Bricken%2C+Brian+Chen%2C+Adam+Pearce%2C+Craig+Citro%2C+Emmanuel+Ameisen%2C+Andy+Jones%2C+Hoagy+Cunningham%2C+Nicholas+L+Turner%2C+Callum+McDougall%2C+)
291. Constantin Venhoff, Iván Arcuschin, Philip Torr, Arthur Conmy, and Neel Nanda. 2025a. Understanding reasoning in thinking language models via steering vectors. Preprint, arXiv:2506.18167. — [arXiv:2506.18167](https://arxiv.org/abs/2506.18167) [find](https://scholar.google.com/scholar?q=Constantin+Venhoff%2C+Iv%C3%A1n+Arcuschin%2C+Philip+Torr%2C+Arthur+Conmy%2C+and+Neel+Nanda%2E+2025a%2E+Understanding+reasoning+in+thinking+language+models+via+steering+vectors%2E+Preprint%2C+arXiv%3A2506%2E18167%2E+%E2%80%94+%5BarXiv%3A250)
292. Constantin Venhoff, Ashkan Khakzar, Sonia Joseph, Philip Torr, and Neel Nanda. 2025b. How visual representations map to language feature space in multimodal llms. arXiv preprint arXiv:2506.11976. — [arXiv:2506.11976](https://arxiv.org/abs/2506.11976) [find](https://scholar.google.com/scholar?q=Constantin+Venhoff%2C+Ashkan+Khakzar%2C+Sonia+Joseph%2C+Philip+Torr%2C+and+Neel+Nanda%2E+2025b%2E+How+visual+representations+map+to+language+feature+space+in+multimodal+llms%2E+arXiv+preprint+arXiv%3A2506%2E11976%2E+%E2%80%94+%5Ba)
293. Jesse Vig, Sebastian Gehrmann, Yonatan Belinkov, Sharon Qian, Daniel Nevo, Yaron Singer, and Stuart Shieber. 2020. Investigating gender bias in language models using causal mediation analysis. In Advances in Neural Information Processing Systems, volume 33, pages 12388–12401. Curran Associates, Inc. [find](https://scholar.google.com/scholar?q=Jesse+Vig%2C+Sebastian+Gehrmann%2C+Yonatan+Belinkov%2C+Sharon+Qian%2C+Daniel+Nevo%2C+Yaron+Singer%2C+and+Stuart+Shieber%2E+2020%2E+Investigating+gender+bias+in+language+models+using+causal+mediation+analysis%2E+In+Adva)
294. Elena Voita, Javier Ferrando, and Christoforos Nalmpantis. 2024. Neurons in large language models: Dead, n-gram, positional. In Findings of the Association for Computational Linguistics: ACL 2024, pages 1288–1301. [find](https://scholar.google.com/scholar?q=Elena+Voita%2C+Javier+Ferrando%2C+and+Christoforos+Nalmpantis%2E+2024%2E+Neurons+in+large+language+models%3A+Dead%2C+n%2Dgram%2C+positional%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+ACL+2024%2C+pag)
295. Elena Voita, David Talbot, Fedor Moiseev, Rico Sennrich, and Ivan Titov. 2019. [Analyzing multi-head self-attention: Specialized heads do the heavy lifting, the rest can be pruned](https://md.research.elamur.ai/free/en/papers/voita-2019-analyzing-multi-head-self-attention/). In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5797–5808.
296. Boshi Wang, Xiang Yue, Yu Su, and Huan Sun. 2024a. Grokking of implicit reasoning in transformers: A mechanistic journey to the edge of generalization. Advances in Neural Information Processing Systems, 37:95238–95265. [find](https://scholar.google.com/scholar?q=Boshi+Wang%2C+Xiang+Yue%2C+Yu+Su%2C+and+Huan+Sun%2E+2024a%2E+Grokking+of+implicit+reasoning+in+transformers%3A+A+mechanistic+journey+to+the+edge+of+generalization%2E+Advances+in+Neural+Information+Processing+System)
297. Haoyu Wang, Yaqing Wang, Tianci Liu, Tuo Zhao, and Jing Gao. 2023a. HadSkip: Homotopic and adaptive layer skipping of pre-trained language models for efficient inference. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 4283–4294, Singapore. Association for Computation… [find](https://scholar.google.com/scholar?q=Haoyu+Wang%2C+Yaqing+Wang%2C+Tianci+Liu%2C+Tuo+Zhao%2C+and+Jing+Gao%2E+2023a%2E+HadSkip%3A+Homotopic+and+adaptive+layer+skipping+of+pre%2Dtrained+language+models+for+efficient+inference%2E+In+Findings+of+the+Associatio)
298. Kevin Ro Wang, Alexandre Variengien, Arthur Conmy, Buck Shlegeris, and Jacob Steinhardt. 2023b. Interpretability in the wild: a circuit for indirect object identification in GPT-2 small. In The Eleventh International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Kevin+Ro+Wang%2C+Alexandre+Variengien%2C+Arthur+Conmy%2C+Buck+Shlegeris%2C+and+Jacob+Steinhardt%2E+2023b%2E+Interpretability+in+the+wild%3A+a+circuit+for+indirect+object+identification+in+GPT%2D2+small%2E+In+The+Eleven)
299. Mengru Wang, Xingyu Chen, Yue Wang, Zhiwei He, Jiahao Xu, Tian Liang, Qiuzhi Liu, Yunzhi Yao, Wenxuan Wang, Ruotian Ma, Haitao Mi, Ningyu Zhang, Zhaopeng Tu, Xiaolong Li, and Dong Yu. 2025a. Two experts are all you need for steering thinking: Reinforcing cognitive effort in moe reasoning models with… [find](https://scholar.google.com/scholar?q=Mengru+Wang%2C+Xingyu+Chen%2C+Yue+Wang%2C+Zhiwei+He%2C+Jiahao+Xu%2C+Tian+Liang%2C+Qiuzhi+Liu%2C+Yunzhi+Yao%2C+Wenxuan+Wang%2C+Ruotian+Ma%2C+Haitao+Mi%2C+Ningyu+Zhang%2C+Zhaopeng+Tu%2C+Xiaolong+Li%2C+and+Dong+Yu%2E+2025a%2E+Two+exper)
300. Mengru Wang, Ziwen Xu, Shengyu Mao, Shumin Deng, Zhaopeng Tu, Huajun Chen, and Ningyu Zhang. 2025b. [Beyond prompt engineering: Robust behavior control in llms via steering target atoms](https://md.research.elamur.ai/free/en/papers/wang-2025-beyond-prompt-engineering-robust-behavior/). arXiv preprint arXiv:2505.20322.
301. Mingyang Wang, Heike Adel, Lukas Lange, Yihong Liu, Ercong Nie, Jannik Strötgen, and Hinrich Schuetze. 2025c. [Lost in multilinguality: Dissecting cross-lingual factual inconsistency in transformer language models](https://md.research.elamur.ai/free/en/papers/wang-2025-lost-multilinguality-dissecting-cross-lingual/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguis…
302. Mingyang Wang, Lukas Lange, Heike Adel, Yunpu Ma, Jannik Strötgen, and Hinrich Schuetze. 2025d. [Language mixing in reasoning language models: Patterns, impact, and internal causes](https://md.research.elamur.ai/free/en/papers/wang-2025-language-mixing-reasoning-language-models/). In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 2637–2665, Suzhou, Chi…
303. Mingze Wang, Ruoxi Yu, Lei Wu, and 1 others. 2024b. How transformers implement induction heads: Approximation and optimization analysis. arXiv eprints, pages arXiv–2410. [find](https://scholar.google.com/scholar?q=Mingze+Wang%2C+Ruoxi+Yu%2C+Lei+Wu%2C+and+1+others%2E+2024b%2E+How+transformers+implement+induction+heads%3A+Approximation+and+optimization+analysis%2E+arXiv+eprints%2C+pages+arXiv%E2%80%932410%2E)
304. Runyu Wang, Peng Ping, Zhengyu Guo, Xiaoye Zhang, Quan Shi, Liting Zhou, and Tianbo Ji. 2025e. Loki: Low-damage knowledge implanting of large language models. Preprint, arXiv:2505.22120. — [arXiv:2505.22120](https://arxiv.org/abs/2505.22120) [find](https://scholar.google.com/scholar?q=Runyu+Wang%2C+Peng+Ping%2C+Zhengyu+Guo%2C+Xiaoye+Zhang%2C+Quan+Shi%2C+Liting+Zhou%2C+and+Tianbo+Ji%2E+2025e%2E+Loki%3A+Low%2Ddamage+knowledge+implanting+of+large+language+models%2E+Preprint%2C+arXiv%3A2505%2E22120%2E+%E2%80%94+%5BarXiv%3A2505)
305. Sizhe Wang, Yongqi Tong, Hengyuan Zhang, Dawei Li, Xin Zhang, and Tianlong Chen. 2024c. Bpo: Towards balanced preference optimization between knowledge breadth and depth in alignment. arXiv preprint arXiv:2411.10914. [find](https://scholar.google.com/scholar?q=Sizhe+Wang%2C+Yongqi+Tong%2C+Hengyuan+Zhang%2C+Dawei+Li%2C+Xin+Zhang%2C+and+Tianlong+Chen%2E+2024c%2E+Bpo%3A+Towards+balanced+preference+optimization+between+knowledge+breadth+and+depth+in+alignment%2E+arXiv+preprint+a)
306. Xinpeng Wang, Chengzhi Hu, Paul Röttger, and Barbara Plank. 2025f. Surgical, cheap, and flexible: Mitigating false refusal in language models via single vector ablation. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Xinpeng+Wang%2C+Chengzhi+Hu%2C+Paul+R%C3%B6ttger%2C+and+Barbara+Plank%2E+2025f%2E+Surgical%2C+cheap%2C+and+flexible%3A+Mitigating+false+refusal+in+language+models+via+single+vector+ablation%2E+In+The+Thirteenth+Internationa)
307. Xinpeng Wang, Mingyang Wang, Yihong Liu, Hinrich Schuetze, and Barbara Plank. 2025g. Refusal direction is universal across safety-aligned languages. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Xinpeng+Wang%2C+Mingyang+Wang%2C+Yihong+Liu%2C+Hinrich+Schuetze%2C+and+Barbara+Plank%2E+2025g%2E+Refusal+direction+is+universal+across+safety%2Daligned+languages%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Inf)
308. Yongjie Wang, Tong Zhang, Xu Guo, and Zhiqi Shen. 2024d. Gradient based feature attribution in explainable AI: A technical review. CoRR, abs/2403.10415. — [arXiv:2403.10415](https://arxiv.org/abs/2403.10415) [find](https://scholar.google.com/scholar?q=Yongjie+Wang%2C+Tong+Zhang%2C+Xu+Guo%2C+and+Zhiqi+Shen%2E+2024d%2E+Gradient+based+feature+attribution+in+explainable+AI%3A+A+technical+review%2E+CoRR%2C+abs%2F2403%2E10415%2E+%E2%80%94+%5BarXiv%3A2403%2E10415%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2403)
309. Zhenyu Wang. 2025. Logitlens4llms: Extending logit lens analysis to modern large language models. arXiv preprint arXiv:2503.11667. — [arXiv:2503.11667](https://arxiv.org/abs/2503.11667) [find](https://scholar.google.com/scholar?q=Zhenyu+Wang%2E+2025%2E+Logitlens4llms%3A+Extending+logit+lens+analysis+to+modern+large+language+models%2E+arXiv+preprint+arXiv%3A2503%2E11667%2E+%E2%80%94+%5BarXiv%3A2503%2E11667%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2503%2E11667%29)
310. Zijian Wang, Yanxiang Ma, and Chang Xu. 2025h. Eliciting chain-of-thought in base llms via gradient-based representation optimization. Preprint, arXiv:2511.19131. — [arXiv:2511.19131](https://arxiv.org/abs/2511.19131) [find](https://scholar.google.com/scholar?q=Zijian+Wang%2C+Yanxiang+Ma%2C+and+Chang+Xu%2E+2025h%2E+Eliciting+chain%2Dof%2Dthought+in+base+llms+via+gradient%2Dbased+representation+optimization%2E+Preprint%2C+arXiv%3A2511%2E19131%2E+%E2%80%94+%5BarXiv%3A2511%2E19131%5D%28https%3A%2F%2Farxiv%2Eor)
311. Ziqi Wang, Hanlin Zhang, Xiner Li, Kuan-Hao Huang, Chi Han, Shuiwang Ji, Sham M. Kakade, Hao Peng, and Heng Ji. 2025i. Eliminating Position Bias of Language Models: A Mechanistic Approach. Preprint, arXiv:2407.01100. — [arXiv:2407.01100](https://arxiv.org/abs/2407.01100) [find](https://scholar.google.com/scholar?q=Ziqi+Wang%2C+Hanlin+Zhang%2C+Xiner+Li%2C+Kuan%2DHao+Huang%2C+Chi+Han%2C+Shuiwang+Ji%2C+Sham+M%2E+Kakade%2C+Hao+Peng%2C+and+Heng+Ji%2E+2025i%2E+Eliminating+Position+Bias+of+Language+Models%3A+A+Mechanistic+Approach%2E+Preprint%2C+a)
312. Jake Ward, Chuqiao Lin, Constantin Venhoff, and Neel Nanda. 2025. Reasoning-finetuning repurposes latent representations in base models. CoRR, abs/2507.12638. — [arXiv:2507.12638](https://arxiv.org/abs/2507.12638) [find](https://scholar.google.com/scholar?q=Jake+Ward%2C+Chuqiao+Lin%2C+Constantin+Venhoff%2C+and+Neel+Nanda%2E+2025%2E+Reasoning%2Dfinetuning+repurposes+latent+representations+in+base+models%2E+CoRR%2C+abs%2F2507%2E12638%2E+%E2%80%94+%5BarXiv%3A2507%2E12638%5D%28https%3A%2F%2Farxiv%2Eorg%2Fab)
313. Chris Wendler, Veniamin Veselovsky, Giovanni Monea, and Robert West. 2024. [Do llamas work in English? on the latent language of multilingual transformers](https://md.research.elamur.ai/free/en/papers/wendler-2024-do-llamas-work-english-latent/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15366–15394, Bangkok, T…
314. Jiaqi Weng, Han Zheng, Hanyu Zhang, Qinqin He, Jialing Tao, Hui Xue, Zhixuan Chu, and Xiting Wang. 2025. [Safe-sail: Towards a fine-grained safety landscape of large language models via sparse autoencoder interpretation framework](https://md.research.elamur.ai/free/en/papers/weng-2026-safe-sail-fine-grained-safety/). arXiv preprint arXiv:2509.18127.
315. Skyler Wu, Eric Meng Shen, Charumathi Badrinath, Jiaqi Ma, and Himabindu Lakkaraju. 2023. [Analyzing chain-of-thought prompting in large language models via gradient-based feature attributions](https://md.research.elamur.ai/free/en/papers/wu-2023-analyzing-chain-thought-prompting-large/). arXiv preprint arXiv:2307.13339.
316. Xuansheng Wu, Jiayi Yuan, Wenlin Yao, Xiaoming Zhai, and Ninghao Liu. 2025a. Interpreting and steering llms with mutual information-based explanations on sparse autoencoders. arXiv preprint arXiv:2502.15576. — [arXiv:2502.15576](https://arxiv.org/abs/2502.15576) [find](https://scholar.google.com/scholar?q=Xuansheng+Wu%2C+Jiayi+Yuan%2C+Wenlin+Yao%2C+Xiaoming+Zhai%2C+and+Ninghao+Liu%2E+2025a%2E+Interpreting+and+steering+llms+with+mutual+information%2Dbased+explanations+on+sparse+autoencoders%2E+arXiv+preprint+arXiv%3A2502)
317. Xuansheng Wu, Haiyan Zhao, Yaochen Zhu, Yucheng Shi, Fan Yang, Lijie Hu, Tianming Liu, Xiaoming Zhai, Wenlin Yao, Jundong Li, and 1 others. 2024a. Usable xai: 10 strategies towards exploiting explainability in the llm era. arXiv preprint arXiv:2403.08946. — [arXiv:2403.08946](https://arxiv.org/abs/2403.08946) [find](https://scholar.google.com/scholar?q=Xuansheng+Wu%2C+Haiyan+Zhao%2C+Yaochen+Zhu%2C+Yucheng+Shi%2C+Fan+Yang%2C+Lijie+Hu%2C+Tianming+Liu%2C+Xiaoming+Zhai%2C+Wenlin+Yao%2C+Jundong+Li%2C+and+1+others%2E+2024a%2E+Usable+xai%3A+10+strategies+towards+exploiting+explaina)
318. Zhaofeng Wu, Xinyan Velocity Yu, Dani Yogatama, Jiasen Lu, and Yoon Kim. 2025b. The semantic hub hypothesis: Language models share semantic representations across languages and modalities. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Zhaofeng+Wu%2C+Xinyan+Velocity+Yu%2C+Dani+Yogatama%2C+Jiasen+Lu%2C+and+Yoon+Kim%2E+2025b%2E+The+semantic+hub+hypothesis%3A+Language+models+share+semantic+representations+across+languages+and+modalities%2E+In+The+Thir)
319. Zhengxuan Wu, Aryaman Arora, Atticus Geiger, Zheng Wang, Jing Huang, Dan Jurafsky, Christopher D Manning, and Christopher Potts. 2025c. Axbench: Steering LLMs? even simple baselines outperform sparse autoencoders. In Forty-second International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Zhengxuan+Wu%2C+Aryaman+Arora%2C+Atticus+Geiger%2C+Zheng+Wang%2C+Jing+Huang%2C+Dan+Jurafsky%2C+Christopher+D+Manning%2C+and+Christopher+Potts%2E+2025c%2E+Axbench%3A+Steering+LLMs%3F+even+simple+baselines+outperform+sparse+)
320. Zhengxuan Wu, Aryaman Arora, Zheng Wang, Atticus Geiger, Dan Jurafsky, Christopher D. Manning, and Christopher Potts. 2024b. Reft: Representation fine-tuning for language models. In Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024,… [find](https://scholar.google.com/scholar?q=Zhengxuan+Wu%2C+Aryaman+Arora%2C+Zheng+Wang%2C+Atticus+Geiger%2C+Dan+Jurafsky%2C+Christopher+D%2E+Manning%2C+and+Christopher+Potts%2E+2024b%2E+Reft%3A+Representation+fine%2Dtuning+for+language+models%2E+In+Advances+in+Neural)
321. Dirk U Wulff and Rui Mata. 2025. Advancing cognitive science with llms. arXiv preprint arXiv:2511.00206. xAI. 2025. Grok 3 beta — the age of reasoning agents. Blog post announcing Grok 3 Beta, describing improvements in reasoning capabilities and performance benchmarks. — [arXiv:2511.00206](https://arxiv.org/abs/2511.00206) [find](https://scholar.google.com/scholar?q=Dirk+U+Wulff+and+Rui+Mata%2E+2025%2E+Advancing+cognitive+science+with+llms%2E+arXiv+preprint+arXiv%3A2511%2E00206%2E+xAI%2E+2025%2E+Grok+3+beta+%E2%80%94+the+age+of+reasoning+agents%2E+Blog+post+announcing+Grok+3+Beta%2C+describ)
322. Heming Xia, Chak Tou Leong, Wenjie Wang, Yongqi Li, and Wenjie Li. 2025. [TokenSkip: Controllable chain-of-thought compression in LLMs](https://md.research.elamur.ai/free/en/papers/xia-2025-tokenskip-controllable-chain-thought-compression/). In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 3351–3363, Suzhou, China. Association for Computational Linguistics.
323. Guangxuan Xiao, Ji Lin, Mickael Seznec, Hao Wu, Julien Demouth, and Song Han. 2023a. Smoothquant: Accurate and efficient post-training quantization for large language models. In International conference on machine learning, pages 38087– 38099. PMLR. [find](https://scholar.google.com/scholar?q=Guangxuan+Xiao%2C+Ji+Lin%2C+Mickael+Seznec%2C+Hao+Wu%2C+Julien+Demouth%2C+and+Song+Han%2E+2023a%2E+Smoothquant%3A+Accurate+and+efficient+post%2Dtraining+quantization+for+large+language+models%2E+In+International+conferen)
324. Guangxuan Xiao, Jiaming Tang, Jingwei Zuo, Junxian Guo, Shang Yang, Haotian Tang, Yao Fu, and Song Han. 2024. [Duoattention: Efficient long-context llm inference with retrieval and streaming heads](https://md.research.elamur.ai/free/en/papers/xiao-2024-duoattention-efficient-long-context-llm/). arXiv preprint arXiv:2410.10819.
325. Guangxuan Xiao, Yuandong Tian, Beidi Chen, Song Han, and Mike Lewis. 2023b. Efficient streaming language models with attention sinks. arXiv. [find](https://scholar.google.com/scholar?q=Guangxuan+Xiao%2C+Yuandong+Tian%2C+Beidi+Chen%2C+Song+Han%2C+and+Mike+Lewis%2E+2023b%2E+Efficient+streaming+language+models+with+attention+sinks%2E+arXiv%2E)
326. Hanqi Xiao, Yi-Lin Sung, Elias Stengel-Eskin, and Mohit Bansal. 2025a. Task-circuit quantization: Leveraging knowledge localization and interpretability for compression. In Second Conference on Language Modeling. [find](https://scholar.google.com/scholar?q=Hanqi+Xiao%2C+Yi%2DLin+Sung%2C+Elias+Stengel%2DEskin%2C+and+Mohit+Bansal%2E+2025a%2E+Task%2Dcircuit+quantization%3A+Leveraging+knowledge+localization+and+interpretability+for+compression%2E+In+Second+Conference+on+Langua)
327. He Xiao, Qingyao Yang, Dirui Xie, Wendong Xu, Wenyong Zhou, Haobo Liu, Zhengwu Liu, and Ngai Wong. 2025b. [Exploring layer-wise information effectiveness for post-training quantization in small language models](https://md.research.elamur.ai/free/en/papers/xiao-2026-exploring-layer-wise-information-effectiveness/). arXiv preprint arXiv:2508.03332.
328. Tinghao Xie, Xiangyu Qi, Yi Zeng, Yangsibo Huang, Udari Madhushani Sehwag, Kaixuan Huang, Luxi He, Boyi Wei, Dacheng Li, Ying Sheng, Ruoxi Jia, Bo Li, Kai Li, Danqi Chen, Peter Henderson, and Prateek Mittal. 2025. [SORRY-Bench: Systematically evaluating large language model safety refusal](https://md.research.elamur.ai/free/en/papers/xie-2024-sorry-bench-systematically-evaluating-large/). In Interna…
329. Wanying Xie, Yang Feng, Shuhao Gu, and Dong Yu. 2021. Importance-based neuron allocation for multilingual neural machine translation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing… [find](https://scholar.google.com/scholar?q=Wanying+Xie%2C+Yang+Feng%2C+Shuhao+Gu%2C+and+Dong+Yu%2E+2021%2E+Importance%2Dbased+neuron+allocation+for+multilingual+neural+machine+translation%2E+In+Proceedings+of+the+59th+Annual+Meeting+of+the+Association+for+C)
330. Jing Xiong, Liyang Fan, Hui Shen, Zunhai Su, Min Yang, Lingpeng Kong, and Ngai Wong. 2025. Dope: Denoising rotary position embedding. arXiv preprint arXiv:2511.09146. — [arXiv:2511.09146](https://arxiv.org/abs/2511.09146) [find](https://scholar.google.com/scholar?q=Jing+Xiong%2C+Liyang+Fan%2C+Hui+Shen%2C+Zunhai+Su%2C+Min+Yang%2C+Lingpeng+Kong%2C+and+Ngai+Wong%2E+2025%2E+Dope%3A+Denoising+rotary+position+embedding%2E+arXiv+preprint+arXiv%3A2511%2E09146%2E+%E2%80%94+%5BarXiv%3A2511%2E09146%5D%28https%3A%2F%2Farxi)
331. Jing Xiong, Qi Han, Yunta Hsieh, Hui Shen, Huajian Xin, Chaofan Tao, Chenyang Zhao, Hengyuan Zhang, Taiqiang Wu, Zhen Zhang, and 1 others. 2026a. Mm-formalizer: Multimodal autoformalization in the wild. arXiv preprint arXiv:2601.03017. — [arXiv:2601.03017](https://arxiv.org/abs/2601.03017) [find](https://scholar.google.com/scholar?q=Jing+Xiong%2C+Qi+Han%2C+Yunta+Hsieh%2C+Hui+Shen%2C+Huajian+Xin%2C+Chaofan+Tao%2C+Chenyang+Zhao%2C+Hengyuan+Zhang%2C+Taiqiang+Wu%2C+Zhen+Zhang%2C+and+1+others%2E+2026a%2E+Mm%2Dformalizer%3A+Multimodal+autoformalization+in+the+wil)
332. Jing Xiong, Chengming Li, Min Yang, Xiping Hu, and Bin Hu. 2026b. [Expression syntax information bottleneck for math word problems](https://md.research.elamur.ai/free/en/papers/xiong-2023-expression-syntax-information-bottleneck-math/). Preprint, arXiv:2310.15664.
333. Jing Xiong, Hui Shen, Shansan Gong, Yuxin Cheng, Jianghan Shen, Chaofan Tao, Haochen Tan, Haoli Bai, Lifeng Shang, and Ngai Wong. 2026c. Ovd: On-policy verbal distillation. arXiv preprint arXiv:2601.21968. — [arXiv:2601.21968](https://arxiv.org/abs/2601.21968) [find](https://scholar.google.com/scholar?q=Jing+Xiong%2C+Hui+Shen%2C+Shansan+Gong%2C+Yuxin+Cheng%2C+Jianghan+Shen%2C+Chaofan+Tao%2C+Haochen+Tan%2C+Haoli+Bai%2C+Lifeng+Shang%2C+and+Ngai+Wong%2E+2026c%2E+Ovd%3A+On%2Dpolicy+verbal+distillation%2E+arXiv+preprint+arXiv%3A2601%2E2)
334. Haoyun Xu, Runzhe Zhan, Yingpeng Ma, Derek F. Wong, and Lidia S. Chao. 2025a. [Let’s focus on neuron: Neuron-level supervised fine-tuning for large language model](https://md.research.elamur.ai/free/en/papers/xu-2025-let-s-focus-neuron-neuron/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 9393–9406, Abu Dhabi, UAE. Association for Comput…
335. Ziwen Xu, Shuxun Wang, Kewei Xu, Haoming Xu, Mengru Wang, Xinle Deng, Yunzhi Yao, Guozhou Zheng, Huajun Chen, and Ningyu Zhang. 2025b. Easyedit2: An easy-to-use steering framework for editing large language models. arXiv preprint arXiv:2504.15133. [find](https://scholar.google.com/scholar?q=Ziwen+Xu%2C+Shuxun+Wang%2C+Kewei+Xu%2C+Haoming+Xu%2C+Mengru+Wang%2C+Xinle+Deng%2C+Yunzhi+Yao%2C+Guozhou+Zheng%2C+Huajun+Chen%2C+and+Ningyu+Zhang%2E+2025b%2E+Easyedit2%3A+An+easy%2Dto%2Duse+steering+framework+for+editing+large+la)
336. Prateek Yadav, Derek Tam, Leshem Choshen, Colin A. Raffel, and Mohit Bansal. 2023. [Ties-merging: Resolving interference when merging models](https://md.research.elamur.ai/free/en/papers/yadav-2023-ties-merging-resolving-interference-merging/). In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, D…
337. Hao Yang, Qianghua Zhao, and Lei Li. 2024. Chain-of-thought in large language models: Decoding, projection, and activation. Preprint, arXiv:2412.03944. — [arXiv:2412.03944](https://arxiv.org/abs/2412.03944) [find](https://scholar.google.com/scholar?q=Hao+Yang%2C+Qianghua+Zhao%2C+and+Lei+Li%2E+2024%2E+Chain%2Dof%2Dthought+in+large+language+models%3A+Decoding%2C+projection%2C+and+activation%2E+Preprint%2C+arXiv%3A2412%2E03944%2E+%E2%80%94+%5BarXiv%3A2412%2E03944%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2412%2E)
338. Jiayu Yang, Yuxuan Fan, Songning Lai, Shengen Wu, Jiaqi Tang, Chun Kang, Zhijiang Guo, and Yutao Yue. 2025a. Ace: Attribution-controlled knowledge editing for multi-hop factual recall. arXiv preprint arXiv:2510.07896. — [arXiv:2510.07896](https://arxiv.org/abs/2510.07896) [find](https://scholar.google.com/scholar?q=Jiayu+Yang%2C+Yuxuan+Fan%2C+Songning+Lai%2C+Shengen+Wu%2C+Jiaqi+Tang%2C+Chun+Kang%2C+Zhijiang+Guo%2C+and+Yutao+Yue%2E+2025a%2E+Ace%3A+Attribution%2Dcontrolled+knowledge+editing+for+multi%2Dhop+factual+recall%2E+arXiv+preprint+)
339. Wanli Yang, Fei Sun, Rui Tang, Hongyu Zang, Du Su, Qi Cao, Jingang Wang, Huawei Shen, and Xueqi Cheng. 2025b. Fine-tuning done right in model editing. arXiv preprint arXiv:2509.22072. — [arXiv:2509.22072](https://arxiv.org/abs/2509.22072) [find](https://scholar.google.com/scholar?q=Wanli+Yang%2C+Fei+Sun%2C+Rui+Tang%2C+Hongyu+Zang%2C+Du+Su%2C+Qi+Cao%2C+Jingang+Wang%2C+Huawei+Shen%2C+and+Xueqi+Cheng%2E+2025b%2E+Fine%2Dtuning+done+right+in+model+editing%2E+arXiv+preprint+arXiv%3A2509%2E22072%2E+%E2%80%94+%5BarXiv%3A2509%2E22)
340. Yunzhi Yao, Ningyu Zhang, Zekun Xi, Mengru Wang, Ziwen Xu, Shumin Deng, and Huajun Chen. 2024a. Knowledge circuits in pretrained transformers. In Advances in Neural Information Processing Systems, volume 37, pages 118571–118602. Curran Associates, Inc. [find](https://scholar.google.com/scholar?q=Yunzhi+Yao%2C+Ningyu+Zhang%2C+Zekun+Xi%2C+Mengru+Wang%2C+Ziwen+Xu%2C+Shumin+Deng%2C+and+Huajun+Chen%2E+2024a%2E+Knowledge+circuits+in+pretrained+transformers%2E+In+Advances+in+Neural+Information+Processing+Systems%2C+vol)
341. Yunzhi Yao, Ningyu Zhang, Zekun Xi, Mengru Wang, Ziwen Xu, Shumin Deng, and Huajun Chen. 2024b. Knowledge circuits in pretrained transformers. Advances in Neural Information Processing Systems, 37:118571–118602. [find](https://scholar.google.com/scholar?q=Yunzhi+Yao%2C+Ningyu+Zhang%2C+Zekun+Xi%2C+Mengru+Wang%2C+Ziwen+Xu%2C+Shumin+Deng%2C+and+Huajun+Chen%2E+2024b%2E+Knowledge+circuits+in+pretrained+transformers%2E+Advances+in+Neural+Information+Processing+Systems%2C+37%3A118)
342. Tian Ye, Zicheng Xu, Yuanzhi Li, and Zeyuan Allen-Zhu. 2025a. [Physics of language models: Part 2.1, grade-school math and the hidden reasoning process](https://md.research.elamur.ai/free/en/papers/ye-2024-physics-language-models-part-2/). In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net.
343. Weihao Ye, Qiong Wu, Wenhao Lin, and Yiyi Zhou. 2025b. Fit and prune: Fast and training-free visual token pruning for multi-modal large language models. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 22128–22136. [find](https://scholar.google.com/scholar?q=Weihao+Ye%2C+Qiong+Wu%2C+Wenhao+Lin%2C+and+Yiyi+Zhou%2E+2025b%2E+Fit+and+prune%3A+Fast+and+training%2Dfree+visual+token+pruning+for+multi%2Dmodal+large+language+models%2E+In+Proceedings+of+the+AAAI+Conference+on+Artifi)
344. Wei Jie Yeo, Nirmalendu Prakash, Clement Neo, Ranjan Satapathy, Roy Ka-Wei Lee, and Erik Cambria. 2025a. Understanding refusal in language models with sparse autoencoders. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 6377–6399, Suzhou, China. Association for Comput… [find](https://scholar.google.com/scholar?q=Wei+Jie+Yeo%2C+Nirmalendu+Prakash%2C+Clement+Neo%2C+Ranjan+Satapathy%2C+Roy+Ka%2DWei+Lee%2C+and+Erik+Cambria%2E+2025a%2E+Understanding+refusal+in+language+models+with+sparse+autoencoders%2E+In+Findings+of+the+Associati)
345. Wei Jie Yeo, Ranjan Satapathy, and Erik Cambria. 2025b. Towards faithful natural language explanations: A study using activation patching in large language models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 10436–10458. [find](https://scholar.google.com/scholar?q=Wei+Jie+Yeo%2C+Ranjan+Satapathy%2C+and+Erik+Cambria%2E+2025b%2E+Towards+faithful+natural+language+explanations%3A+A+study+using+activation+patching+in+large+language+models%2E+In+Proceedings+of+the+2025+Conferenc)
346. Fangcong Yin, Xi Ye, and Greg Durrett. 2024. Lofit: Localized fine-tuning on llm representations. Advances in Neural Information Processing Systems, 37:9474–9506. [find](https://scholar.google.com/scholar?q=Fangcong+Yin%2C+Xi+Ye%2C+and+Greg+Durrett%2E+2024%2E+Lofit%3A+Localized+fine%2Dtuning+on+llm+representations%2E+Advances+in+Neural+Information+Processing+Systems%2C+37%3A9474%E2%80%939506%2E)
347. Kayo Yin and Graham Neubig. 2022. Interpreting language models with contrastive explanations. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 184–198, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Kayo+Yin+and+Graham+Neubig%2E+2022%2E+Interpreting+language+models+with+contrastive+explanations%2E+In+Proceedings+of+the+2022+Conference+on+Empirical+Methods+in+Natural+Language+Processing%2C+pages+184%E2%80%93198%2C+)
348. Qingyu Yin, Chak Tou Leong, Linyi Yang, Wenxuan Huang, Wenjie Li, Xiting Wang, Jaehong Yoon, YunXing, XingYu, and Jinjin Gu. 2025. Refusal falls off a cliff: How safety alignment fails in reasoning? Preprint, arXiv:2510.06036. — [arXiv:2510.06036](https://arxiv.org/abs/2510.06036) [find](https://scholar.google.com/scholar?q=Qingyu+Yin%2C+Chak+Tou+Leong%2C+Linyi+Yang%2C+Wenxuan+Huang%2C+Wenjie+Li%2C+Xiting+Wang%2C+Jaehong+Yoon%2C+YunXing%2C+XingYu%2C+and+Jinjin+Gu%2E+2025%2E+Refusal+falls+off+a+cliff%3A+How+safety+alignment+fails+in+reasoning%3F+P)
349. Weiqiu You, Anton Xue, Shreya Havaldar, Delip Rao, Helen Jin, Chris Callison-Burch, and Eric Wong. 2025. Probabilistic soundness guarantees in llm reasoning chains. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 7517–7536, Suzhou, China. Association… [find](https://scholar.google.com/scholar?q=Weiqiu+You%2C+Anton+Xue%2C+Shreya+Havaldar%2C+Delip+Rao%2C+Helen+Jin%2C+Chris+Callison%2DBurch%2C+and+Eric+Wong%2E+2025%2E+Probabilistic+soundness+guarantees+in+llm+reasoning+chains%2E+In+Proceedings+of+the+2025+Conferen)
350. Haeun Yu, Seogyeong Jeong, Siddhesh Pawar, Jisu Shin, Jiho Jin, Junho Myung, Alice Oh, and Isabelle Augenstein. 2025a. Entangled in Representations: Mechanistic Investigation of Cultural Biases in Large Language Models. Preprint, arXiv:2508.08879. [find](https://scholar.google.com/scholar?q=Haeun+Yu%2C+Seogyeong+Jeong%2C+Siddhesh+Pawar%2C+Jisu+Shin%2C+Jiho+Jin%2C+Junho+Myung%2C+Alice+Oh%2C+and+Isabelle+Augenstein%2E+2025a%2E+Entangled+in+Representations%3A+Mechanistic+Investigation+of+Cultural+Biases+in+Lar)
351. Mengxia Yu, De Wang, Qi Shan, Colorado J Reed, and Alvin Wan. 2024. The super weight in large language models. arXiv preprint arXiv:2411.07191. — [arXiv:2411.07191](https://arxiv.org/abs/2411.07191) [find](https://scholar.google.com/scholar?q=Mengxia+Yu%2C+De+Wang%2C+Qi+Shan%2C+Colorado+J+Reed%2C+and+Alvin+Wan%2E+2024%2E+The+super+weight+in+large+language+models%2E+arXiv+preprint+arXiv%3A2411%2E07191%2E+%E2%80%94+%5BarXiv%3A2411%2E07191%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2411%2E07191%29)
352. Yijiong Yu, Huiqiang Jiang, Xufang Luo, Qianhui Wu, Chin-Yew Lin, Dongsheng Li, Yuqing Yang, Yongfeng Huang, and Lili Qiu. 2025b. Mitigate position bias in LLMs via scaling a single hidden states channel. In Findings of the Association for Computational Linguistics: ACL 2025, pages 6092–6111, Vienna… [find](https://scholar.google.com/scholar?q=Yijiong+Yu%2C+Huiqiang+Jiang%2C+Xufang+Luo%2C+Qianhui+Wu%2C+Chin%2DYew+Lin%2C+Dongsheng+Li%2C+Yuqing+Yang%2C+Yongfeng+Huang%2C+and+Lili+Qiu%2E+2025b%2E+Mitigate+position+bias+in+LLMs+via+scaling+a+single+hidden+states+chan)
353. Zeping Yu and Sophia Ananiadou. 2024a. How do large language models learn in-context? query and key matrices of in-context heads are two towers for metric learning. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA, November 12-16,… [find](https://scholar.google.com/scholar?q=Zeping+Yu+and+Sophia+Ananiadou%2E+2024a%2E+How+do+large+language+models+learn+in%2Dcontext%3F+query+and+key+matrices+of+in%2Dcontext+heads+are+two+towers+for+metric+learning%2E+In+Proceedings+of+the+2024+Conferen)
354. Zeping Yu and Sophia Ananiadou. 2024b. Interpreting arithmetic mechanism in large language models through comparative neuron analysis. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 3293–3306, Miami, Florida, USA. Association for Computational Lingui… [find](https://scholar.google.com/scholar?q=Zeping+Yu+and+Sophia+Ananiadou%2E+2024b%2E+Interpreting+arithmetic+mechanism+in+large+language+models+through+comparative+neuron+analysis%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+Methods+in+Nat)
355. Zeping Yu and Sophia Ananiadou. 2024c. Neuron-level knowledge attribution in large language models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA, November 12-16, 2024, pages 3267–3280. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Zeping+Yu+and+Sophia+Ananiadou%2E+2024c%2E+Neuron%2Dlevel+knowledge+attribution+in+large+language+models%2E+In+Proceedings+of+the+2024+Conference+on+Empirical+Methods+in+Natural+Language+Processing%2C+EMNLP+202)
356. Zeping Yu and Sophia Ananiadou. 2024d. Understanding multimodal llms: the mechanistic interpretability of llava in visual question answering. arXiv preprint arXiv:2411.10950. — [arXiv:2411.10950](https://arxiv.org/abs/2411.10950) [find](https://scholar.google.com/scholar?q=Zeping+Yu+and+Sophia+Ananiadou%2E+2024d%2E+Understanding+multimodal+llms%3A+the+mechanistic+interpretability+of+llava+in+visual+question+answering%2E+arXiv+preprint+arXiv%3A2411%2E10950%2E+%E2%80%94+%5BarXiv%3A2411%2E10950%5D%28http)
357. Zeping Yu and Sophia Ananiadou. 2025. Understanding and Mitigating Gender Bias in LLMs via Interpretable Neuron Editing. Preprint, arXiv:2501.14457. — [arXiv:2501.14457](https://arxiv.org/abs/2501.14457) [find](https://scholar.google.com/scholar?q=Zeping+Yu+and+Sophia+Ananiadou%2E+2025%2E+Understanding+and+Mitigating+Gender+Bias+in+LLMs+via+Interpretable+Neuron+Editing%2E+Preprint%2C+arXiv%3A2501%2E14457%2E+%E2%80%94+%5BarXiv%3A2501%2E14457%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2501%2E144)
358. Zeping Yu, Yonatan Belinkov, and Sophia Ananiadou. 2025c. [Back attention: Understanding and enhancing multi-hop reasoning in large language models](https://md.research.elamur.ai/free/en/papers/yu-2025-back-attention-understanding-enhancing-multi/). In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 11268–11283, Suzhou, China. Association for Computation…
359. Shuzhou Yuan, Zhan Qu, Mario Tawfelis, and Michael Färber. 2025. From monolingual to bilingual: Investigating language conditioning in large language models for psycholinguistic tasks. arXiv preprint arXiv:2508.02502. [find](https://scholar.google.com/scholar?q=Shuzhou+Yuan%2C+Zhan+Qu%2C+Mario+Tawfelis%2C+and+Michael+F%C3%A4rber%2E+2025%2E+From+monolingual+to+bilingual%3A+Investigating+language+conditioning+in+large+language+models+for+psycholinguistic+tasks%2E+arXiv+preprint+)
360. Dharunish Yugeswardeenoo, Harshil Nukala, Cole Blondin, Sean O’Brien, Vasu Sharma, and Kevin Zhu. 2025. Interpreting the latent structure of operator precedence in language models. In The First Workshop on the Interplay of Model Behavior and Model Internals. [find](https://scholar.google.com/scholar?q=Dharunish+Yugeswardeenoo%2C+Harshil+Nukala%2C+Cole+Blondin%2C+Sean+O%E2%80%99Brien%2C+Vasu+Sharma%2C+and+Kevin+Zhu%2E+2025%2E+Interpreting+the+latent+structure+of+operator+precedence+in+language+models%2E+In+The+First+Worksh)
361. Binrui Zeng, Bin Ji, Xiaodong Liu, Jie Yu, Shasha Li, Jun Ma, Xiaopeng Li, Shangwen Wang, Xinran Hong, and Yongtao Tang. 2024. Lsaq: Layer-specific adaptive quantization for large language model deployment. arXiv preprint arXiv:2412.18135. — [arXiv:2412.18135](https://arxiv.org/abs/2412.18135) [find](https://scholar.google.com/scholar?q=Binrui+Zeng%2C+Bin+Ji%2C+Xiaodong+Liu%2C+Jie+Yu%2C+Shasha+Li%2C+Jun+Ma%2C+Xiaopeng+Li%2C+Shangwen+Wang%2C+Xinran+Hong%2C+and+Yongtao+Tang%2E+2024%2E+Lsaq%3A+Layer%2Dspecific+adaptive+quantization+for+large+language+model+deplo)
362. Feng Zhang, Yanbin Liu, Weihua Li, Jie Lv, Xiaodan Wang, and Quan Bai. 2025a. Towards superior quantization accuracy: A layer-sensitive approach. arXiv preprint arXiv:2503.06518. — [arXiv:2503.06518](https://arxiv.org/abs/2503.06518) [find](https://scholar.google.com/scholar?q=Feng+Zhang%2C+Yanbin+Liu%2C+Weihua+Li%2C+Jie+Lv%2C+Xiaodan+Wang%2C+and+Quan+Bai%2E+2025a%2E+Towards+superior+quantization+accuracy%3A+A+layer%2Dsensitive+approach%2E+arXiv+preprint+arXiv%3A2503%2E06518%2E+%E2%80%94+%5BarXiv%3A2503%2E06518%5D%28)
363. Fred Zhang and Neel Nanda. 2023. [Towards best practices of activation patching in language models: Metrics and methods](https://md.research.elamur.ai/free/en/papers/zhang-2023-towards-best-practices-activation-patching/). arXiv preprint arXiv:2309.16042.
364. Hengyuan Zhang, Xinrong Chen, Yingmin Qiu, Xiao Liang, Ziyue Li, Guanyu Wang, Weiping Li, Tong Mo, Hayden Kwok-Hay So, and Ngai Wong. 2025b. Guilomo: Allocating expert number and rank for loramoe via bilevel optimization with guidedselection vectors. Preprint. [find](https://scholar.google.com/scholar?q=Hengyuan+Zhang%2C+Xinrong+Chen%2C+Yingmin+Qiu%2C+Xiao+Liang%2C+Ziyue+Li%2C+Guanyu+Wang%2C+Weiping+Li%2C+Tong+Mo%2C+Hayden+Kwok%2DHay+So%2C+and+Ngai+Wong%2E+2025b%2E+Guilomo%3A+Allocating+expert+number+and+rank+for+loramoe+via+)
365. Hengyuan Zhang, Xinrong Chen, Zunhai Su, Xiao Liang, Jing Xiong, Wendong Xu, He Xiao, Chaofan Tao, Wei Zhang, Ruobing Xie, and 1 others. 2026. Beyond outliers: A data-free layer-wise mixed-precision quantization approach driven by numerical and structural dual-sensitivity. arXiv preprint arXiv:2603.… — [arXiv:2603.17354](https://arxiv.org/abs/2603.17354) [find](https://scholar.google.com/scholar?q=Hengyuan+Zhang%2C+Xinrong+Chen%2C+Zunhai+Su%2C+Xiao+Liang%2C+Jing+Xiong%2C+Wendong+Xu%2C+He+Xiao%2C+Chaofan+Tao%2C+Wei+Zhang%2C+Ruobing+Xie%2C+and+1+others%2E+2026%2E+Beyond+outliers%3A+A+data%2Dfree+layer%2Dwise+mixed%2Dprecision+q)
366. Hengyuan Zhang, Zitao Liu, Chenming Shang, Dawei Li, and Yong Jiang. 2025c. A question-centric multi-experts contrastive learning framework for improving the accuracy and interpretability of deep sequential knowledge tracing models. ACM Transactions on Knowledge Discovery from Data, 19(2):1–25. [find](https://scholar.google.com/scholar?q=Hengyuan+Zhang%2C+Zitao+Liu%2C+Chenming+Shang%2C+Dawei+Li%2C+and+Yong+Jiang%2E+2025c%2E+A+question%2Dcentric+multi%2Dexperts+contrastive+learning+framework+for+improving+the+accuracy+and+interpretability+of+deep+sequ)
367. Hengyuan Zhang, Chenming Shang, Sizhe Wang, Dongdong Zhang, Yiyao Yu, Feng Yao, Renliang Sun, Yujiu Yang, and Furu Wei. 2025d. [ShifCon: Enhancing non-dominant language capabilities with a shift-based multilingual contrastive framework](https://md.research.elamur.ai/free/en/papers/zhang-2025-shifcon-enhancing-non-dominant-language/). In Proceedings of the 63rd Annual Meeting of the Association for…
368. Hengyuan Zhang, Yanru Wu, Dawei Li, Sak Yang, Rui Zhao, Yong Jiang, and Fei Tan. 2024a. Balancing speciality and versatility: a coarse to fine framework for supervised fine-tuning large language model. In Findings of the Association for Computational Linguistics, ACL 2024, Bangkok, Thailand and virt… [find](https://scholar.google.com/scholar?q=Hengyuan+Zhang%2C+Yanru+Wu%2C+Dawei+Li%2C+Sak+Yang%2C+Rui+Zhao%2C+Yong+Jiang%2C+and+Fei+Tan%2E+2024a%2E+Balancing+speciality+and+versatility%3A+a+coarse+to+fine+framework+for+supervised+fine%2Dtuning+large+language+model)
369. Hengyuan Zhang, Shiping Yang, Xiao Liang, Chenming Shang, Yuxuan Jiang, Chaofan Tao, Jing Xiong, Hayden Kwok-Hay So, Ruobing Xie, Angel X Chang, and 1 others. 2025e. [Find your optimal teacher: Personalized data synthesis via router-guided multi-teacher distillation](https://md.research.elamur.ai/free/en/papers/zhang-2026-find-your-optimal-teacher-personalized/). arXiv preprint arXiv:2510.10925.
370. Jason Zhang and Scott Viteri. 2025. Uncovering latent chain of thought vectors in language models. Preprint, arXiv:2409.14026. [find](https://scholar.google.com/scholar?q=Jason+Zhang+and+Scott+Viteri%2E+2025%2E+Uncovering+latent+chain+of+thought+vectors+in+language+models%2E+Preprint%2C+arXiv%3A2409%2E14026%2E)
371. Jiawei Zhang. 2019. Cognitive functions of the brain: perception, attention and memory. arXiv preprint arXiv:1907.02863. — [arXiv:1907.02863](https://arxiv.org/abs/1907.02863) [find](https://scholar.google.com/scholar?q=Jiawei+Zhang%2E+2019%2E+Cognitive+functions+of+the+brain%3A+perception%2C+attention+and+memory%2E+arXiv+preprint+arXiv%3A1907%2E02863%2E+%E2%80%94+%5BarXiv%3A1907%2E02863%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1907%2E02863%29)
372. Ningyu Zhang, Yunzhi Yao, Bozhong Tian, Peng Wang, Shumin Deng, Mengru Wang, Zekun Xi, Shengyu Mao, Jintian Zhang, Yuansheng Ni, Siyuan Cheng, Ziwen Xu, Xin Xu, Jia-Chen Gu, Yong Jiang, Pengjun Xie, Fei Huang, Lei Liang, Zhiqiang Zhang, and 3 others. 2024b. A comprehensive study of knowledge editing… [find](https://scholar.google.com/scholar?q=Ningyu+Zhang%2C+Yunzhi+Yao%2C+Bozhong+Tian%2C+Peng+Wang%2C+Shumin+Deng%2C+Mengru+Wang%2C+Zekun+Xi%2C+Shengyu+Mao%2C+Jintian+Zhang%2C+Yuansheng+Ni%2C+Siyuan+Cheng%2C+Ziwen+Xu%2C+Xin+Xu%2C+Jia%2DChen+Gu%2C+Yong+Jiang%2C+Pengjun+Xie%2C+F)
373. Shaolei Zhang, Tian Yu, and Yang Feng. 2024c. [TruthX: Alleviating hallucinations by editing large language models in truthful space](https://md.research.elamur.ai/free/en/papers/zhang-2024-truthx-alleviating-hallucinations-editing-large/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8908– 8949, Bangkok, Thailand. Association fo…
374. Wei Zhang, Chaoqun Wan, Yonggang Zhang, Yiu ming Cheung, Xinmei Tian, Xu Shen, and Jieping Ye. 2024d. Interpreting and improving large language models in arithmetic calculation. In Forty-first International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=Wei+Zhang%2C+Chaoqun+Wan%2C+Yonggang+Zhang%2C+Yiu+ming+Cheung%2C+Xinmei+Tian%2C+Xu+Shen%2C+and+Jieping+Ye%2E+2024d%2E+Interpreting+and+improving+large+language+models+in+arithmetic+calculation%2E+In+Forty%2Dfirst+Interna)
375. Xue Zhang, Yunlong Liang, Fandong Meng, Songming Zhang, Yufeng Chen, Jinan Xu, and Jie Zhou. 2025f. [Multilingual knowledge editing with language-agnostic factual neurons](https://md.research.elamur.ai/free/en/papers/zhang-2025-multilingual-knowledge-editing-language-agnostic/). In Proceedings of the 31st International Conference on Computational Linguistics, COLING 2025, Abu Dhabi, UAE, January 19-24, 202…
376. Zhenyu Zhang, Xiaoxia Wu, Zhongzhu Zhou, Qingyang Wu, Yineng Zhang, Pragaash Ponnusamy, Harikaran Subbaraj, Jue Wang, Shuaiwen Leon Song, and Ben Athiwaratkun. 2025g. Understanding and steering the cognitive behaviors of reasoning models at test-time. arXiv preprint arXiv:2512.24574. — [arXiv:2512.24574](https://arxiv.org/abs/2512.24574) [find](https://scholar.google.com/scholar?q=Zhenyu+Zhang%2C+Xiaoxia+Wu%2C+Zhongzhu+Zhou%2C+Qingyang+Wu%2C+Yineng+Zhang%2C+Pragaash+Ponnusamy%2C+Harikaran+Subbaraj%2C+Jue+Wang%2C+Shuaiwen+Leon+Song%2C+and+Ben+Athiwaratkun%2E+2025g%2E+Understanding+and+steering+the+co)
377. Zhenyu Zhang, Shujian Zhang, John Lambert, Wenxuan Zhou, Zhangyang Wang, Mingqing Chen, Andrew Hard, Rajiv Mathews, and Lun Wang. 2025h. Fantastic reasoning behaviors and where to find them: Unsupervised discovery of the reasoning process. arXiv preprint arXiv:2512.23988. — [arXiv:2512.23988](https://arxiv.org/abs/2512.23988) [find](https://scholar.google.com/scholar?q=Zhenyu+Zhang%2C+Shujian+Zhang%2C+John+Lambert%2C+Wenxuan+Zhou%2C+Zhangyang+Wang%2C+Mingqing+Chen%2C+Andrew+Hard%2C+Rajiv+Mathews%2C+and+Lun+Wang%2E+2025h%2E+Fantastic+reasoning+behaviors+and+where+to+find+them%3A+Unsupervi)
378. Zhi Zhang, Srishti Yadav, Fengze Han, and Ekaterina Shutova. 2025i. Cross-modal information flow in multimodal large language models. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2025, Nashville, TN, USA, June 11-15, 2025, pages 19781–19791. Computer Vision Foundation / IE… [find](https://scholar.google.com/scholar?q=Zhi+Zhang%2C+Srishti+Yadav%2C+Fengze+Han%2C+and+Ekaterina+Shutova%2E+2025i%2E+Cross%2Dmodal+information+flow+in+multimodal+large+language+models%2E+In+IEEE%2FCVF+Conference+on+Computer+Vision+and+Pattern+Recognition%2C)
379. Zhihao Zhang, Qiaole Dong, Qi Zhang, Jun Zhao, Enyu Zhou, Zhiheng Xi, Senjie Jin, Xiaoran Fan, Yuhao Zhou, Mingqi Wu, Yanwei Fu, Tao Ji, Tao Gui, Xuanjing Huang, and Kai Chen. 2025j. Why reinforcement fine-tuning enables mllms preserve prior knowledge better: A data perspective. Preprint, arXiv:2506… — [arXiv:2506.23508](https://arxiv.org/abs/2506.23508) [find](https://scholar.google.com/scholar?q=Zhihao+Zhang%2C+Qiaole+Dong%2C+Qi+Zhang%2C+Jun+Zhao%2C+Enyu+Zhou%2C+Zhiheng+Xi%2C+Senjie+Jin%2C+Xiaoran+Fan%2C+Yuhao+Zhou%2C+Mingqi+Wu%2C+Yanwei+Fu%2C+Tao+Ji%2C+Tao+Gui%2C+Xuanjing+Huang%2C+and+Kai+Chen%2E+2025j%2E+Why+reinforcement)
380. Zhihao Zhang, Jun Zhao, Qi Zhang, Tao Gui, and Xuanjing Huang. 2024e. [Unveiling linguistic regions in large language models](https://md.research.elamur.ai/free/en/papers/zhang-2024-unveiling-linguistic-regions-large-language/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Bangkok, Thailand, August 11-16, 2024, pages 6228…
381. Zhong Zhang, Bang Liu, and Junming Shao. 2023. Fine-tuning happens in tiny subspaces: Exploring intrinsic task-specific subspaces of pre-trained language models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1701–1713, Toron… [find](https://scholar.google.com/scholar?q=Zhong+Zhang%2C+Bang+Liu%2C+and+Junming+Shao%2E+2023%2E+Fine%2Dtuning+happens+in+tiny+subspaces%3A+Exploring+intrinsic+task%2Dspecific+subspaces+of+pre%2Dtrained+language+models%2E+In+Proceedings+of+the+61st+Annual+Meet)
382. Delong Zhao, Qiang Huang, Di Yan, Yiqun Sun, and Jun Yu. 2025a. Partially shared concept bottleneck models. arXiv preprint arXiv:2511.22170. — [arXiv:2511.22170](https://arxiv.org/abs/2511.22170) [find](https://scholar.google.com/scholar?q=Delong+Zhao%2C+Qiang+Huang%2C+Di+Yan%2C+Yiqun+Sun%2C+and+Jun+Yu%2E+2025a%2E+Partially+shared+concept+bottleneck+models%2E+arXiv+preprint+arXiv%3A2511%2E22170%2E+%E2%80%94+%5BarXiv%3A2511%2E22170%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2511%2E22170%29)
383. Dezhi Zhao, Xin Liu, Xiaocheng Feng, Hui Wang, and Bing Qin. 2025b. Probing and boosting large language models capabilities via attention heads. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 28518–28532, Suzhou, China. Association for Computational… [find](https://scholar.google.com/scholar?q=Dezhi+Zhao%2C+Xin+Liu%2C+Xiaocheng+Feng%2C+Hui+Wang%2C+and+Bing+Qin%2E+2025b%2E+Probing+and+boosting+large+language+models+capabilities+via+attention+heads%2E+In+Proceedings+of+the+2025+Conference+on+Empirical+Meth)
384. Jiachen Zhao, Jing Huang, Zhengxuan Wu, David Bau, and Weiyan Shi. 2025c. LLMs encode harmfulness and refusal separately. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. [find](https://scholar.google.com/scholar?q=Jiachen+Zhao%2C+Jing+Huang%2C+Zhengxuan+Wu%2C+David+Bau%2C+and+Weiyan+Shi%2E+2025c%2E+LLMs+encode+harmfulness+and+refusal+separately%2E+In+The+Thirty%2Dninth+Annual+Conference+on+Neural+Information+Processing+Systems)
385. Yiran Zhao, Wenxuan Zhang, Guizhen Chen, Kenji Kawaguchi, and Lidong Bing. 2024b. [How do large language models handle multilingualism?](https://md.research.elamur.ai/free/en/papers/zhao-2024-large-language-models-handle-multilingualism/) In Advances in Neural Information Processing Systems, volume 37, pages 15296–15319. Curran Associates, Inc.
386. Yiran Zhao, Wenxuan Zhang, Yuxi Xie, Anirudh Goyal, Kenji Kawaguchi, and Michael Shieh. 2025d. Understanding and enhancing safety mechanisms of LLMs via safety-specific neuron. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Yiran+Zhao%2C+Wenxuan+Zhang%2C+Yuxi+Xie%2C+Anirudh+Goyal%2C+Kenji+Kawaguchi%2C+and+Michael+Shieh%2E+2025d%2E+Understanding+and+enhancing+safety+mechanisms+of+LLMs+via+safety%2Dspecific+neuron%2E+In+The+Thirteenth+Inter)
387. Yu Zhao, Alessio Devoto, Giwon Hong, Xiaotang Du, Aryo Pradipta Gema, Hongru Wang, Xuanli He, Kam-Fai Wong, and Pasquale Minervini. 2025e. Steering knowledge selection behaviours in llms via sae-based representation engineering. In Proceedings of the 2025 Conference of the Nations of the Americas Ch… [find](https://scholar.google.com/scholar?q=Yu+Zhao%2C+Alessio+Devoto%2C+Giwon+Hong%2C+Xiaotang+Du%2C+Aryo+Pradipta+Gema%2C+Hongru+Wang%2C+Xuanli+He%2C+Kam%2DFai+Wong%2C+and+Pasquale+Minervini%2E+2025e%2E+Steering+knowledge+selection+behaviours+in+llms+via+sae%2Dbased)
388. Yu Zhao, Xiaotang Du, Giwon Hong, Aryo Pradipta Gema, Alessio Devoto, Hongru Wang, Xuanli He, Kam-Fai Wong, and Pasquale Minervini. 2024c. Analysing the residual stream of language models under knowledge conflicts. CoRR, abs/2410.16090. — [arXiv:2410.16090](https://arxiv.org/abs/2410.16090) [find](https://scholar.google.com/scholar?q=Yu+Zhao%2C+Xiaotang+Du%2C+Giwon+Hong%2C+Aryo+Pradipta+Gema%2C+Alessio+Devoto%2C+Hongru+Wang%2C+Xuanli+He%2C+Kam%2DFai+Wong%2C+and+Pasquale+Minervini%2E+2024c%2E+Analysing+the+residual+stream+of+language+models+under+knowle)
389. Yanqi Zhou, Tao Lei, Hanxiao Liu, Nan Du, Yanping Huang, Vincent Y. Zhao, Andrew M. Dai, Zhifeng Chen, Quoc V. Le, and James Laudon. 2022. Mixture-of-experts with expert choice routing. In Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing System… [find](https://scholar.google.com/scholar?q=Yanqi+Zhou%2C+Tao+Lei%2C+Hanxiao+Liu%2C+Nan+Du%2C+Yanping+Huang%2C+Vincent+Y%2E+Zhao%2C+Andrew+M%2E+Dai%2C+Zhifeng+Chen%2C+Quoc+V%2E+Le%2C+and+James+Laudon%2E+2022%2E+Mixture%2Dof%2Dexperts+with+expert+choice+routing%2E+In+Advances+in)
390. Zhenhong Zhou, Haiyang Yu, Xinghua Zhang, Rongwu Xu, Fei Huang, Kun Wang, Yang Liu, Junfeng Fang, and Yongbin Li. 2025. On the role of attention heads in large language model safety. In The Thirteenth International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Zhenhong+Zhou%2C+Haiyang+Yu%2C+Xinghua+Zhang%2C+Rongwu+Xu%2C+Fei+Huang%2C+Kun+Wang%2C+Yang+Liu%2C+Junfeng+Fang%2C+and+Yongbin+Li%2E+2025%2E+On+the+role+of+attention+heads+in+large+language+model+safety%2E+In+The+Thirteenth)
391. Shaolin Zhu, Leiyu Pan, Bo Li, and Deyi Xiong. 2024. LANDeRMT: Dectecting and routing language-aware neurons for selectively finetuning LLMs to machine translation. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 12135– 12148,… [find](https://scholar.google.com/scholar?q=Shaolin+Zhu%2C+Leiyu+Pan%2C+Bo+Li%2C+and+Deyi+Xiong%2E+2024%2E+LANDeRMT%3A+Dectecting+and+routing+language%2Daware+neurons+for+selectively+finetuning+LLMs+to+machine+translation%2E+In+Proceedings+of+the+62nd+Annual+M)
392. Haiyan Zhao, Hanjie Chen, Fan Yang, Ninghao Liu, Huiqi Deng, Hengyi Cai, Shuaiqiang Wang, Dawei Yin, and Mengnan Du. 2024a. [Explainability for large language models: A survey](https://md.research.elamur.ai/free/en/papers/zhao-2023-explainability-large-language-models-survey/). ACM Transactions on Intelligent Systems and Technology, 15(2):1–38.

*Updated 2 November 2026*