---
title: "Controllable LLM Reasoning via Sparse Autoencoder-Based"
description: "Large Reasoning Models (LRMs) choose their own reasoning moves, such as re-checking or… 45 references, cited by 3 papers here."
authors: ["Yi Fang", "Wenjie Wang", "Mingfeng Xue", "Boyi Deng", "Fengli Xu", "Dayiheng Liu", "Fuli Feng"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.974/"
doi: "10.18653/v1/2026.acl-long.974"
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/fang-2026-controllable-llm-reasoning-sparse-autoencoder/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/fang-2026-controllable-llm-reasoning-sparse-autoencoder/)

[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/5/) / Controllable LLM Reasoning via Sparse Autoencoder-Based Steering

# Controllable LLM Reasoning via Sparse Autoencoder-Based Steering

**Authors:** [Yi Fang](https://md.research.elamur.ai/free/en/authors/yi-fang/), [Wenjie Wang](https://md.research.elamur.ai/free/en/authors/wenjie-wang/), [Mingfeng Xue](https://md.research.elamur.ai/free/en/authors/mingfeng-xue/), [Boyi Deng](https://md.research.elamur.ai/free/en/authors/boyi-deng/), [Fengli Xu](https://md.research.elamur.ai/free/en/authors/fengli-xu/), [Dayiheng Liu](https://md.research.elamur.ai/free/en/authors/dayiheng-liu/) and [Fuli Feng](https://md.research.elamur.ai/free/en/authors/fuli-feng/) · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/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.acl-long.974/) · [PDF](https://aclanthology.org/2026.acl-long.974.pdf) · [DOI 10.18653/v1/2026.acl-long.974](https://doi.org/10.18653/v1/2026.acl-long.974) · [OpenAlex](https://openalex.org/W7166821494)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [3](https://md.research.elamur.ai/free/en/papers/fang-2026-controllable-llm-reasoning-sparse-autoencoder/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
Large Reasoning Models (LRMs) choose their own reasoning moves, such as re-checking or backtracking, and these choices can lead to wasted effort or wrong answers. The authors aim to let users steer which strategy the model applies next. Prompt instructions are often ignored. Contrastive activation vectors tend to mix several strategies together. To get cleaner directions, the authors train Sparse Autoencoders (SAEs) on a model layer's residual stream. Each SAE feature tends to encode one concept, and a chosen feature is added repeatedly to the hidden state while the model generates subsequent tokens. The SAE has tens of thousands of features, so the proposed SAE-Steering pipeline selects useful ones in two steps. First, a cheap filter keeps only features that raise the logits of strategy keywords. Second, the surviving features are ranked by how well they actually steer generation on a small validation set. The selected features control strategies more than 15% better than existing methods. They can also push a model off a faulty path, which improves accuracy. This matters for making LRM reasoning more reliable and interpretable.

## Key points
- Five strategies are targeted: Problem Understanding, Procedural Planning, Backtracking, Multi-Perspective Verification and Hypothesis Reasoning. The authors describe the method as general beyond these five.
- The SAEs use Top-K sparsity. Steering strength α is lowered step by step from a high value until repetitive output disappears.
- Stage 1 discards over 99% of features, shrinking roughly 30k candidates to fewer than 100. Stage 2 ranks the remainder by measured control effect.
- Reported results:
  - over 15% higher control effectiveness than baselines;
  - 7% absolute accuracy gain when redirecting erroneous reasoning;
  - about 5% average improvement in correction rate.
- Limitations: the pipeline relies on manually chosen strategy keywords and covers only five strategies. Negative steering was explored only briefly.
- Code and data are at github.com/Peter-Fy/SAE-Steering. The paper appears at ACL 2026.

*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

- [Tsinghua University](https://md.research.elamur.ai/free/en/universities/tsinghua-university/) — University, Beijing, China · 333 works · 87,019 citations
- [University of Science and Technology of China](https://md.research.elamur.ai/free/en/universities/university-of-science-and-technology-of-china/) — University, Hefei, China · 108 works · 112,698 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works · [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,426 works

## Cite

Cite as: Yi Fang, Wenjie Wang, Mingfeng Xue, Boyi Deng, Fengli Xu, Dayiheng Liu and Fuli Feng. 2026. Controllable LLM Reasoning via Sparse Autoencoder-Based Steering. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

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

```bibtex
@inproceedings{fang2026controllable,
  title = {{Controllable LLM Reasoning via Sparse Autoencoder-Based Steering}},
  author = {Yi Fang, Wenjie Wang, Mingfeng Xue, Boyi Deng, Fengli Xu, Dayiheng Liu and Fuli Feng},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.974},
  url = {https://doi.org/10.18653/v1/2026.acl-long.974}
}
```

- [← previous Confidence Should Be Calibrated More Than One Turn Deep](https://md.research.elamur.ai/free/en/papers/zhang-2026-confidence-should-be-calibrated-more/)
- [next → Controlling Reading Ease with Gaze-Guided Text Generation](https://md.research.elamur.ai/free/en/papers/sauberli-2026-controlling-reading-ease-gaze-guided/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/5/)

## References

45 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. AIME. 2025. Aime problems and solutions. [find](https://scholar.google.com/scholar?q=AIME%2E+2025%2E+Aime+problems+and+solutions%2E)
2. 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+)
3. Xingyu Chen, Jiahao Xu, Tian Liang, Zhiwei He, Jianhui Pang, Dian Yu, Linfeng Song, Qiuzhi Liu, Mengfei Zhou, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, and Dong Yu. 2025. Do NOT think that much for 2+3=? on the overthinking of long reasoning models. In ICML. OpenReview.net. [find](https://scholar.google.com/scholar?q=Xingyu+Chen%2C+Jiahao+Xu%2C+Tian+Liang%2C+Zhiwei+He%2C+Jianhui+Pang%2C+Dian+Yu%2C+Linfeng+Song%2C+Qiuzhi+Liu%2C+Mengfei+Zhou%2C+Zhuosheng+Zhang%2C+Rui+Wang%2C+Zhaopeng+Tu%2C+Haitao+Mi%2C+and+Dong+Yu%2E+2025%2E+Do+NOT+think+that+mu)
4. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, and 1 others. 2025. [Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities](https://md.research.elamur.ai/free/en/papers/comanici-2025-gemini-2-5-pushing-frontier/). arX…
5. Boyi Deng, Yu Wan, Baosong Yang, Yidan Zhang, and Fuli Feng. 2025. [Unveiling language-specific features in large language models via sparse autoencoders](https://md.research.elamur.ai/free/en/papers/deng-2025-unveiling-language-specific-features-large/). In ACL (1), pages 4563–4608. Association for Computational Linguistics.
6. 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. arXiv pr… [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)
7. Zihao Fu, Wai Lam, Anthony Man-Cho So, and Bei Shi. 2021. A theoretical analysis of the repetition problem in text generation. In AAAI. [find](https://scholar.google.com/scholar?q=Zihao+Fu%2C+Wai+Lam%2C+Anthony+Man%2DCho+So%2C+and+Bei+Shi%2E+2021%2E+A+theoretical+analysis+of+the+repetition+problem+in+text+generation%2E+In+AAAI%2E)
8. Jia LI, Edward Beeching, Lewis Tunstall, Ben Lipkin, Roman Soletskyi, Shengyi Costa Huang, Kashif Rasul, Longhui Yu, Albert Jiang, Ziju Shen, Zihan Qin, Bin Dong, Li Zhou, Yann Fleureau, Guillaume Lample, and Stanislas Polu. 2024. Numinamath. [find](https://scholar.google.com/scholar?q=Jia+LI%2C+Edward+Beeching%2C+Lewis+Tunstall%2C+Ben+Lipkin%2C+Roman+Soletskyi%2C+Shengyi+Costa+Huang%2C+Kashif+Rasul%2C+Longhui+Yu%2C+Albert+Jiang%2C+Ziju+Shen%2C+Zihan+Qin%2C+Bin+Dong%2C+Li+Zhou%2C+Yann+Fleureau%2C+Guillaume+Lam)
9. Hunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. 2023. [Let’s verify step by step](https://md.research.elamur.ai/free/en/papers/lightman-2023-let-s-verify-step-step/). arXiv preprint arXiv:2305.20050.
10. Zhengkai Lin, Zhihang Fu, Ze Chen, Chao Chen, Liang Xie, Wenxiao Wang, Deng Cai, Zheng Wang, and Jieping Ye. 2025. Controlling thinking speed in reasoning models. NeurIPS. [find](https://scholar.google.com/scholar?q=Zhengkai+Lin%2C+Zhihang+Fu%2C+Ze+Chen%2C+Chao+Chen%2C+Liang+Xie%2C+Wenxiao+Wang%2C+Deng+Cai%2C+Zheng+Wang%2C+and+Jieping+Ye%2E+2025%2E+Controlling+thinking+speed+in+reasoning+models%2E+NeurIPS%2E)
11. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. [find](https://scholar.google.com/scholar?q=Aixin+Liu%2C+Bei+Feng%2C+Bing+Xue%2C+Bingxuan+Wang%2C+Bochao+Wu%2C+Chengda+Lu%2C+Chenggang+Zhao%2C+Chengqi+Deng%2C+Chenyu+Zhang%2C+Chong+Ruan%2C+and+1+others%2E+2024%2E+Deepseek%2Dv3+technical+report%2E+arXiv+preprint+arXiv%3A2412)
12. Andrey V. Galichin, Alexey Dontsov, Polina Druzhinina, Anton Razzhigaev, Oleg Y. Rogov, Elena Tutubalina, and Ivan V. 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+V%2E+Galichin%2C+Alexey+Dontsov%2C+Polina+Druzhinina%2C+Anton+Razzhigaev%2C+Oleg+Y%2E+Rogov%2C+Elena+Tutubalina%2C+and+Ivan+V%2E+Oseledets%2E+2025%2E+I+have+covered+all+the+bases+here%3A+Interpreting+reasoning+feature)
13. Sara Vera Marjanović, Arkil Patel, Vaibhav Adlakha, Milad Aghajohari, Parishad BehnamGhader, Mehar Bhatia, Aditi Khandelwal, Austin Kraft, Benno Krojer, Xing Han Lù, Nicholas Meade, Dongchan Shin, Amirhossein Kazemnejad, Gaurav Kamath, Marius Mosbach, Karolina Stańczak, and Siva Reddy. 2025. Deeps… [find](https://scholar.google.com/scholar?q=Sara+Vera+Marjanovic%CC%81%2C+Arkil+Patel%2C+Vaibhav+Adlakha%2C+Milad+Aghajohari%2C+Parishad+BehnamGhader%2C+Mehar+Bhatia%2C+Aditi+Khandelwal%2C+Austin+Kraft%2C+Benno+Krojer%2C+Xing+Han+L%C3%B9%2C+Nicholas+Meade%2C+Dongchan+Shin%2C+Am)
14. Kanishk Gandhi, Ayush Chakravarthy, Anikait Singh, Nathan Lile, and Noah D. Goodman. 2025. [Cognitive behaviors that enable self-improving reasoners, or, four habits of highly effective stars](https://md.research.elamur.ai/free/en/papers/gandhi-2025-cognitive-behaviors-that-enable-self/). Second Conference on Language Modeling.
15. Niklas Muennighoff, Zitong Yang, Weijia Shi, Xiang Lisa Li, Li Fei-Fei, Hannaneh Hajishirzi, Luke Zettlemoyer, Percy Liang, Emmanuel J. Candès, and Tatsunori Hashimoto. 2025. [s1: Simple test-time scaling](https://md.research.elamur.ai/free/en/papers/muennighoff-2025-s1-simple-test-time-scaling/). arXiv preprint arXiv:2501.19393.
16. 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). nostalgebraist. 2020. Interpreting gpt: The logit lens. Less-Wrong (blog). [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)
17. Leo Gao, Tom Dupré la Tour, Henk Tillman, Gabriel Goh, Rajan Troll, Alec Radford, Ilya Sutskever, Jan Leike, and Jeffrey Wu. 2025. Scaling and evaluating sparse autoencoders. In ICLR. [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+2025%2E+Scaling+and+evaluating+sparse+autoencoders%2E+In+ICLR%2E)
18. OpenAI. 2025. [Openai o1 system card](https://md.research.elamur.ai/free/en/papers/openai-2024-openai-o1-system-card/). Accessed: 2025-02-21.
19. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, and 1 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/). arXiv preprint arXiv:2501.12948.
20. OpenAI. 2024. [Gpt-4o system card](https://md.research.elamur.ai/free/en/papers/openai-2024-gpt-4o-system-card/). arXiv:2410.21276.
21. Preprint, Qianjun Pan, Wenkai Ji, Yuyang Ding, Junsong Li, Shil-ian Chen, Junyi Wang, Jie Zhou, Qin Chen, Min Zhang, Yulan Wu, and Liang He. 2025. A survey of slow thinking-based reasoning llms using reinforced learning and inference-time scaling law. arXiv preprint arXiv:2505.02665. [find](https://scholar.google.com/scholar?q=Preprint%2C+Qianjun+Pan%2C+Wenkai+Ji%2C+Yuyang+Ding%2C+Junsong+Li%2C+Shil%2Dian+Chen%2C+Junyi+Wang%2C+Jie+Zhou%2C+Qin+Chen%2C+Min+Zhang%2C+Yulan+Wu%2C+and+Liang+He%2E+2025%2E+A+survey+of+slow+thinking%2Dbased+reasoning+llms+using+)
22. Robert Huben, Hoagy Cunningham, Logan Riggs Smith, Aidan Ewart, and Lee Sharkey. 2024. Sparse autoencoders find highly interpretable features in language models. In ICLR. [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+ICLR%2E)
23. Yunjia Qi, Hao Peng, Xiaozhi Wang, Amy Xin, Youfeng Liu, Bin Xu, Lei Hou, and Juanzi Li. 2025. AGEN-TIF: benchmarking instruction following of large language models in agentic scenarios. CoRR, abs/2505.16944. — [arXiv:2505.16944](https://arxiv.org/abs/2505.16944) [find](https://scholar.google.com/scholar?q=Yunjia+Qi%2C+Hao+Peng%2C+Xiaozhi+Wang%2C+Amy+Xin%2C+Youfeng+Liu%2C+Bin+Xu%2C+Lei+Hou%2C+and+Juanzi+Li%2E+2025%2E+AGEN%2DTIF%3A+benchmarking+instruction+following+of+large+language+models+in+agentic+scenarios%2E+CoRR%2C+abs%2F250)
24. Vladimir Karpukhin, Barlas Oguz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. 2020. [Dense passage retrieval for open-domain question answering](https://md.research.elamur.ai/free/en/papers/karpukhin-2020-dense-passage-retrieval-open-domain/). In EMNLP (1). Association for Computational Linguistics.
25. David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R. Bowman. 2023. [GPQA: A graduate-level google-proof q&a benchmark](https://md.research.elamur.ai/free/en/papers/rein-2023-gpqa-graduate-level-google-proof/). CoRR, abs/2311.12022.
26. Leheng Sheng, An Zhang, Zijian Wu, Weixiang Zhao, Changshuo Shen, Yi Zhang, Xiang Wang, and Tat-Seng Chua. 2025. On reasoning strength planning in large reasoning models. NeurIPS. [find](https://scholar.google.com/scholar?q=Leheng+Sheng%2C+An+Zhang%2C+Zijian+Wu%2C+Weixiang+Zhao%2C+Changshuo+Shen%2C+Yi+Zhang%2C+Xiang+Wang%2C+and+Tat%2DSeng+Chua%2E+2025%2E+On+reasoning+strength+planning+in+large+reasoning+models%2E+NeurIPS%2E)
27. Wei Shi, Sihang Li, Tao Liang, Mingyang Wan, Gojun Ma, Xiang Wang, and Xiangnan He. 2025. [Route sparse autoencoder to interpret large language models](https://md.research.elamur.ai/free/en/papers/shi-2025-route-sparse-autoencoder-interpret-large/). CoRR, abs/2503.08200.
28. Noah Shinn, Federico Cassano, Ashwin Gopinath, Karthik Narasimhan, and Shunyu Yao. 2023. [Reflexion: language agents with verbal reinforcement learning](https://md.research.elamur.ai/free/en/papers/shinn-2023-reflexion-language-agents-verbal-reinforcement/). In NeurIPS.
29. Chandan Singh, Jeevana Priya Inala, Michel Galley, Rich Caruana, and Jianfeng Gao. 2024. [Rethinking interpretability in the era of large language models](https://md.research.elamur.ai/free/en/papers/singh-2024-rethinking-interpretability-era-large-language/). arXiv preprint arXiv:2402.01761.
30. Charlie Victor Snell, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. 2025. Scaling LLM test-time compute optimally can be more effective than scaling parameters for reasoning. In ICLR. OpenReview.net. [find](https://scholar.google.com/scholar?q=Charlie+Victor+Snell%2C+Jaehoon+Lee%2C+Kelvin+Xu%2C+and+Aviral+Kumar%2E+2025%2E+Scaling+LLM+test%2Dtime+compute+optimally+can+be+more+effective+than+scaling+parameters+for+reasoning%2E+In+ICLR%2E+OpenReview%2Enet%2E)
31. Xinyu Tang, Xiaolei Wang, Zhihao Lv, Yingqian Min, 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 ACL (1), pages 6832– 6849. Association for Computational Linguistics.
32. Openthoughts Team. 2025a. [Openthoughts: Data recipes for reasoning models](https://md.research.elamur.ai/free/en/papers/guha-2025-openthoughts-data-recipes-reasoning-models/).
33. Qwen Team. 2025b. Qwen3 technical report. [find](https://scholar.google.com/scholar?q=Qwen+Team%2E+2025b%2E+Qwen3+technical+report%2E)
34. Aäron van den Oord, Yazhe Li, and Oriol Vinyals. 2018. [Representation learning with contrastive predictive coding](https://md.research.elamur.ai/free/en/papers/oord-2018-representation-learning-contrastive-predictive-coding/). arXiv preprint arXiv:1807.03748.
35. Zeyu Yun, Yubei Chen, Bruno A. Olshausen, and Yann LeCun. 2021. Transformer visualization via dictionary learning: contextualized embedding as a linear superposition of transformer factors. In DeeLIO@NAACL-HLT, pages 1–10. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Zeyu+Yun%2C+Yubei+Chen%2C+Bruno+A%2E+Olshausen%2C+and+Yann+LeCun%2E+2021%2E+Transformer+visualization+via+dictionary+learning%3A+contextualized+embedding+as+a+linear+superposition+of+transformer+factors%2E+In+DeeLIO%40)
36. Wojciech Zaremba, Evgenia Nitishinskaya, Boaz Barak, Stephanie Lin, Sam Toyer, Yaodong Yu, Rachel Dias, Eric Wallace, Kai Xiao, Johannes Heidecke, and 1 others. 2025. Trading inference-time compute for adversarial robustness. arXiv preprint arXiv:2501.18841. [find](https://scholar.google.com/scholar?q=Wojciech+Zaremba%2C+Evgenia+Nitishinskaya%2C+Boaz+Barak%2C+Stephanie+Lin%2C+Sam+Toyer%2C+Yaodong+Yu%2C+Rachel+Dias%2C+Eric+Wallace%2C+Kai+Xiao%2C+Johannes+Heidecke%2C+and+1+others%2E+2025%2E+Trading+inference%2Dtime+compute+fo)
37. Xingsheng Zhang, Luxi Xing, Chen Zhang, Yanbing Liu, Yifan Deng, Yunpeng Li, Yue Hu, and Chenxu Niu. 2025. Can we steer reasoning direction by thinking intervention? In Findings of EMNLP 2025. [find](https://scholar.google.com/scholar?q=Xingsheng+Zhang%2C+Luxi+Xing%2C+Chen+Zhang%2C+Yanbing+Liu%2C+Yifan+Deng%2C+Yunpeng+Li%2C+Yue+Hu%2C+and+Chenxu+Niu%2E+2025%2E+Can+we+steer+reasoning+direction+by+thinking+intervention%3F+In+Findings+of+EMNLP+2025%2E)
38. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Tianle Li, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zhuohan Li, Zi Lin, Eric P. Xing, Joseph E. Gonzalez, Ion Stoica, and Hao Zhang. 2024. [Lmsys-chat-1m: A large-scale real-world LLM conversation dataset](https://md.research.elamur.ai/free/en/papers/zheng-2023-lmsys-chat-1m-large-scale/). In ICLR. OpenReview.net.
39. Qihuang Zhong, Kang Wang, Ziyang Xu, Juhua Liu, Liang Ding, and Bo Du. 2024. Achieving\> 97% on gsm8k: Deeply understanding the problems makes llms better solvers for math word problems. Frontiers of Computer Science. [find](https://scholar.google.com/scholar?q=Qihuang+Zhong%2C+Kang+Wang%2C+Ziyang+Xu%2C+Juhua+Liu%2C+Liang+Ding%2C+and+Bo+Du%2E+2024%2E+Achieving%5C%3E+97%25+on+gsm8k%3A+Deeply+understanding+the+problems+makes+llms+better+solvers+for+math+word+problems%2E+Frontiers+of+)
40. Pei Zhou, Jay Pujara, Xiang Ren, Xinyun Chen, Heng-Tze Cheng, Quoc V. Le, Ed H. Chi, Denny Zhou, Swaroop Mishra, and Huaixiu Steven Zheng. 2024. SELF-DISCOVER: large language models self-compose reasoning structures. In NeurIPS. [find](https://scholar.google.com/scholar?q=Pei+Zhou%2C+Jay+Pujara%2C+Xiang+Ren%2C+Xinyun+Chen%2C+Heng%2DTze+Cheng%2C+Quoc+V%2E+Le%2C+Ed+H%2E+Chi%2C+Denny+Zhou%2C+Swaroop+Mishra%2C+and+Huaixiu+Steven+Zheng%2E+2024%2E+SELF%2DDISCOVER%3A+large+language+models+self%2Dcompose+reaso)
41. Constantin Venhoff, Iván Arcuschin, Philip Torr, Arthur Conmy, and Neel Nanda. 2025. Understanding reasoning in thinking language models via steering vectors. CoRR, abs/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+2025%2E+Understanding+reasoning+in+thinking+language+models+via+steering+vectors%2E+CoRR%2C+abs%2F2506%2E18167%2E+%E2%80%94+%5BarXiv%3A2506%2E18167)
42. Yue Wang, Qiuzhi Liu, Jiahao Xu, Tian Liang, Xingyu Chen, Zhiwei He, Linfeng Song, Dian Yu, Juntao Li, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, and Dong Yu. 2025. Thoughts are all over the place: On the underthinking of o1-like llms. arXiv preprint arXiv:2501.18585. [find](https://scholar.google.com/scholar?q=Yue+Wang%2C+Qiuzhi+Liu%2C+Jiahao+Xu%2C+Tian+Liang%2C+Xingyu+Chen%2C+Zhiwei+He%2C+Linfeng+Song%2C+Dian+Yu%2C+Juntao+Li%2C+Zhuosheng+Zhang%2C+Rui+Wang%2C+Zhaopeng+Tu%2C+Haitao+Mi%2C+and+Dong+Yu%2E+2025%2E+Thoughts+are+all+over+the+p)
43. Tong Wu, Chong Xiang, Jiachen T. Wang, and Prateek Mittal. 2025. Effectively controlling reasoning models through thinking intervention. arXiv preprint arXiv:2503.24370. — [arXiv:2503.24370](https://arxiv.org/abs/2503.24370) [find](https://scholar.google.com/scholar?q=Tong+Wu%2C+Chong+Xiang%2C+Jiachen+T%2E+Wang%2C+and+Prateek+Mittal%2E+2025%2E+Effectively+controlling+reasoning+models+through+thinking+intervention%2E+arXiv+preprint+arXiv%3A2503%2E24370%2E+%E2%80%94+%5BarXiv%3A2503%2E24370%5D%28https%3A%2F%2Fa)
44. Chenxu Yang, Qingyi Si, Mz Dai, Dingyu Yao, Mingyu Zheng, Minghui Chen, Zheng Lin, and Weiping Wang. 2025a. Test-time prompt intervention. [find](https://scholar.google.com/scholar?q=Chenxu+Yang%2C+Qingyi+Si%2C+Mz+Dai%2C+Dingyu+Yao%2C+Mingyu+Zheng%2C+Minghui+Chen%2C+Zheng+Lin%2C+and+Weiping+Wang%2E+2025a%2E+Test%2Dtime+prompt+intervention%2E)
45. Jingyuan Yang, Rongjun Li, Weixuan Wang, Ziyu Zhou, Zhiyong Feng, and Wei Peng. 2025b. Lf-steering: Latent feature activation steering for enhancing semantic consistency in large language models. arXiv preprint arXiv:2501.11036. — [arXiv:2501.11036](https://arxiv.org/abs/2501.11036) [find](https://scholar.google.com/scholar?q=Jingyuan+Yang%2C+Rongjun+Li%2C+Weixuan+Wang%2C+Ziyu+Zhou%2C+Zhiyong+Feng%2C+and+Wei+Peng%2E+2025b%2E+Lf%2Dsteering%3A+Latent+feature+activation+steering+for+enhancing+semantic+consistency+in+large+language+models%2E+arXi)

*Updated 31 October 2026*