---
title: "CoDA — paper by Shi et al. (2026) — summary and ideas"
description: "It offers a training-free alternative to retraining or model editing. 59 references. Among them: 0."
authors: ["JinWei Shi", "Qizhuo Xie", "Qianzi Hou", "Zhipeng Wang", "Wanting Su", "Jianhua Zhao", "Tao Zheng", "Tieke He"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.576/"
doi: "10.18653/v1/2026.findings-acl.576"
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/shi-2026-coda-restoring-contextual-dominance-copy/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/shi-2026-coda-restoring-contextual-dominance-copy/)

[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/25/) / CoDA: Restoring Contextual Dominance via Copy-Encouraged Attention Intervention for Mitigating RAG Hallucinations

# CoDA: Restoring Contextual Dominance via Copy-Encouraged Attention Intervention for Mitigating RAG Hallucinations

**Authors:** JinWei Shi, Qizhuo Xie, Qianzi Hou, Zhipeng Wang, Wanting Su, Jianhua Zhao, Tao Zheng and Tieke He · **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.576/) · [PDF](https://aclanthology.org/2026.findings-acl.576.pdf) · [DOI 10.18653/v1/2026.findings-acl.576](https://doi.org/10.18653/v1/2026.findings-acl.576) · [OpenAlex](https://openalex.org/W7166870810)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
The paper examines a specific RAG failure: the retrieved passage contains correct and sufficient evidence, yet the model still answers from its parametric memory. The authors treat this as a routing problem in the residual stream. Building on the ReDeEP decomposition, they measure two quantities per layer:
- an External Context Score (ES), the cosine alignment between a generated token and the context tokens it attends to most;
- a Parametric Knowledge Score (PS), the Jensen–Shannon divergence of LogitLens distributions before and after each FFN.

Hallucinated and factual tokens separate in mid-to-late layers, where attention becomes less selective toward the context and FFN-driven memory takes over. CoDA works offline and then at inference time. Offline, it identifies these hallucination-prone layers. At inference, it selects top-attended context anchors and boosts value states that are semantically consistent with the evidence. This makes ordinary heads behave more like copying heads, without changing any weights. On Llama-family models, CoDA improves contextual faithfulness, reduces hallucination, and stays robust to long, noisy contexts at moderate latency cost. It offers a training-free alternative to retraining or model editing.

## Key points
- **Models:** experiments use Llama2-7B, Llama2-13B and Llama3-8B. The authors acknowledge that other architectures are untested.
- **Benchmarks:** RAGTruth, Dolly (AC) and six CoFaithfulQA subsets (including NQ and TriviaQA). The metrics include Context Recall (ConR) and memory reliance (MemR).
- **Baselines:** Vanilla-RAG, ReDeEP, SFT, ParamMute, KAFT, C-DPO, COIECD and prompting methods.
- **Noise robustness:** in a controlled test, gold evidence is always present while unrelated passages are added to vary length and noise. CoDA shows the smallest ConR drop on all six CoFaithfulQA subsets.
- **Latency:** overhead is roughly 15–17% per sample. For example, Llama3-8B on RAGTruth goes from 7.82 s to 9.13 s. Layer localization is done offline, and the overhead stays steady as context length grows.

*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

- [Nanjing University](https://md.research.elamur.ai/free/en/universities/nanjing-university/) — University, Nanjing, China · 74 works · 7,814 citations

📚 Shelves (5): [🖥️ 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 · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works

## Cite

Cite as: JinWei Shi, Qizhuo Xie, Qianzi Hou, Zhipeng Wang, Wanting Su, Jianhua Zhao, Tao Zheng and Tieke He. 2026. CoDA: Restoring Contextual Dominance via Copy-Encouraged Attention Intervention for Mitigating RAG Hallucinations. 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{shi2026coda,
  title = {{CoDA: Restoring Contextual Dominance via Copy-Encouraged Attention Intervention for Mitigating RAG Hallucinations}},
  author = {JinWei Shi, Qizhuo Xie, Qianzi Hou, Zhipeng Wang, Wanting Su, Jianhua Zhao, Tao Zheng and Tieke He},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.576},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.576}
}
```

- [← previous CobwebTM: Probabilistic Concept Formation for Lifelong and Hierarchical Topic Modeling](https://md.research.elamur.ai/free/en/papers/singaravadivelan-2026-cobwebtm-probabilistic-concept-formation-lifelong/)
- [next → CodecMOS-Accent: A MOS Benchmark of Resynthesized and TTS Speech from Neural Codecs Across English Accents](https://md.research.elamur.ai/free/en/papers/huang-2026-codecmos-accent-mos-benchmark-resynthesized/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/25/)

## References

59 references, 24 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi. 2023. [Self-rag: Learning to retrieve, generate, and critique through self-reflection](https://md.research.elamur.ai/free/en/papers/asai-2023-self-rag-learning-retrieve-generate/). Preprint, arXiv:2310.11511.
2. Loris Bergeron, Ioana Buhnila, Jérôme François, and Radu State. 2025. Halluguard: Evidence-grounded small reasoning models to mitigate hallucinations in retrieval-augmented generation. Preprint, arXiv:2510.00880. [find](https://scholar.google.com/scholar?q=Loris+Bergeron%2C+Ioana+Buhnila%2C+J%C3%A9r%C3%B4me+Fran%C3%A7ois%2C+and+Radu+State%2E+2025%2E+Halluguard%3A+Evidence%2Dgrounded+small+reasoning+models+to+mitigate+hallucinations+in+retrieval%2Daugmented+generation%2E+Preprint%2C+arXiv)
3. Baolong Bi, Shaohan Huang, Yiwei Wang, Tianchi Yang, Zihan Zhang, Haizhen Huang, Lingrui Mei, Junfeng Fang, Zehao Li, Furu Wei, Weiwei Deng, Feng Sun, Qi Zhang, and Shenghua Liu. 2024. Context-dpo: Aligning language models for context-faithfulness. Preprint, arXiv:2412.15280. [find](https://scholar.google.com/scholar?q=Baolong+Bi%2C+Shaohan+Huang%2C+Yiwei+Wang%2C+Tianchi+Yang%2C+Zihan+Zhang%2C+Haizhen+Huang%2C+Lingrui+Mei%2C+Junfeng+Fang%2C+Zehao+Li%2C+Furu+Wei%2C+Weiwei+Deng%2C+Feng+Sun%2C+Qi+Zhang%2C+and+Shenghua+Liu%2E+2024%2E+Context%2Ddpo%3A+Al)
4. Baolong Bi, Shenghua Liu, Yiwei Wang, Yilong Xu, Junfeng Fang, Lingrui Mei, and Xueqi Cheng. 2025. [find](https://scholar.google.com/scholar?q=Baolong+Bi%2C+Shenghua+Liu%2C+Yiwei+Wang%2C+Yilong+Xu%2C+Junfeng+Fang%2C+Lingrui+Mei%2C+and+Xueqi+Cheng%2E+2025%2E)
5. Parameters vs. context: Fine-grained control of knowledge reliance in language models. Preprint, arXiv:2503.15888. — [arXiv:2503.15888](https://arxiv.org/abs/2503.15888) [find](https://scholar.google.com/scholar?q=Parameters+vs%2E+context%3A+Fine%2Dgrained+control+of+knowledge+reliance+in+language+models%2E+Preprint%2C+arXiv%3A2503%2E15888%2E+%E2%80%94+%5BarXiv%3A2503%2E15888%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2503%2E15888%29)
6. Baiyu Chen, Wilson Wongso, Xiaoqian Hu, Yue Tan, and Flora Salim. 2025. Multi-stage verification-centric framework for mitigating hallucination in multi-modal rag. Preprint, arXiv:2507.20136. — [arXiv:2507.20136](https://arxiv.org/abs/2507.20136) [find](https://scholar.google.com/scholar?q=Baiyu+Chen%2C+Wilson+Wongso%2C+Xiaoqian+Hu%2C+Yue+Tan%2C+and+Flora+Salim%2E+2025%2E+Multi%2Dstage+verification%2Dcentric+framework+for+mitigating+hallucination+in+multi%2Dmodal+rag%2E+Preprint%2C+arXiv%3A2507%2E20136%2E+%E2%80%94+%5BarXiv)
7. Chao Chen, Kai Liu, Ze Chen, Yi Gu, Yue Wu, Mingyuan Tao, Zhihang Fu, and Jieping Ye. 2024. Inside: Llms’ internal states retain the power of hallucination detection. Preprint, arXiv:2402.03744. [find](https://scholar.google.com/scholar?q=Chao+Chen%2C+Kai+Liu%2C+Ze+Chen%2C+Yi+Gu%2C+Yue+Wu%2C+Mingyuan+Tao%2C+Zhihang+Fu%2C+and+Jieping+Ye%2E+2024%2E+Inside%3A+Llms%E2%80%99+internal+states+retain+the+power+of+hallucination+detection%2E+Preprint%2C+arXiv%3A2402%2E03744%2E)
8. Yung-Sung Chuang, Yujia Xie, Hongyin Luo, Yoon Kim, James 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/). Preprint, arXiv:2309.03883.
9. Damai Dai, Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, and Furu Wei. 2022a. Knowledge neurons in pretrained transformers. Preprint, arXiv:2104.08696. [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+2022a%2E+Knowledge+neurons+in+pretrained+transformers%2E+Preprint%2C+arXiv%3A2104%2E08696%2E)
10. Damai Dai, Wenbin Jiang, Qingxiu Dong, Yajuan Lyu, Qiaoqiao She, and Zhifang Sui. 2022b. Neural knowledge bank for pretrained transformers. Preprint, arXiv:2208.00399. — [arXiv:2208.00399](https://arxiv.org/abs/2208.00399) [find](https://scholar.google.com/scholar?q=Damai+Dai%2C+Wenbin+Jiang%2C+Qingxiu+Dong%2C+Yajuan+Lyu%2C+Qiaoqiao+She%2C+and+Zhifang+Sui%2E+2022b%2E+Neural+knowledge+bank+for+pretrained+transformers%2E+Preprint%2C+arXiv%3A2208%2E00399%2E+%E2%80%94+%5BarXiv%3A2208%2E00399%5D%28https%3A%2F%2Farx)
11. Anupam Datta, Matt Fredrikson, Klas Leino, Kaiji Lu, Shayak Sen, Ricardo Shih, and Zifan Wang. 2022. Exploring conceptual soundness with trulens. In NeurIPS 2021 Competitions and Demonstrations Track, pages 302–307. PMLR. [find](https://scholar.google.com/scholar?q=Anupam+Datta%2C+Matt+Fredrikson%2C+Klas+Leino%2C+Kaiji+Lu%2C+Shayak+Sen%2C+Ricardo+Shih%2C+and+Zifan+Wang%2E+2022%2E+Exploring+conceptual+soundness+with+trulens%2E+In+NeurIPS+2021+Competitions+and+Demonstrations+Track%2C)
12. 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)
13. Wenqi Fan, Yujuan Ding, Liangbo Ning, Shijie Wang, Hengyun Li, Dawei Yin, Tat-Seng Chua, and Qing Li. 2024. [A survey on rag meeting llms: Towards retrieval-augmented large language models](https://md.research.elamur.ai/free/en/papers/fan-2024-survey-rag-meeting-llms-towards/). Preprint, arXiv:2405.06211.
14. Robert Friel and Atindriyo Sanyal. 2023. Chainpoll: A high efficacy method for llm hallucination detection. Preprint, arXiv:2310.18344. — [arXiv:2310.18344](https://arxiv.org/abs/2310.18344) [find](https://scholar.google.com/scholar?q=Robert+Friel+and+Atindriyo+Sanyal%2E+2023%2E+Chainpoll%3A+A+high+efficacy+method+for+llm+hallucination+detection%2E+Preprint%2C+arXiv%3A2310%2E18344%2E+%E2%80%94+%5BarXiv%3A2310%2E18344%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2310%2E18344%29)
15. Yunfan Gao, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yi Dai, Jiawei Sun, Meng Wang, and Haofen Wang. 2024. [Retrieval-augmented generation for large language models: A survey](https://md.research.elamur.ai/free/en/papers/gao-2023-retrieval-augmented-generation-large-language/). Preprint, arXiv:2312.10997.
16. Danying Ge, Jianhua Gao, Yixue Yang, and Weixing Ji. 2025. Ha-rag: Hotness-aware rag acceleration via mixed precision and data placement. Preprint, arXiv:2510.20878. — [arXiv:2510.20878](https://arxiv.org/abs/2510.20878) [find](https://scholar.google.com/scholar?q=Danying+Ge%2C+Jianhua+Gao%2C+Yixue+Yang%2C+and+Weixing+Ji%2E+2025%2E+Ha%2Drag%3A+Hotness%2Daware+rag+acceleration+via+mixed+precision+and+data+placement%2E+Preprint%2C+arXiv%3A2510%2E20878%2E+%E2%80%94+%5BarXiv%3A2510%2E20878%5D%28https%3A%2F%2Farxiv)
17. Mor Geva, Avi Caciularu, Kevin Ro 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/). Preprint, arXiv:2203.14680.
18. Mor Geva, Roei Schuster, Jonathan Berant, and Omer Levy. 2021. Transformer feed-forward layers are key-value memories. Preprint, arXiv:2012.14913. [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+Preprint%2C+arXiv%3A2012%2E14913%2E)
19. Sachin Goyal, Christina Baek, J. Zico Kolter, and Aditi Raghunathan. 2025. Context-parametric inversion: Why instruction finetuning can worsen context reliance. Preprint, arXiv:2410.10796. — [arXiv:2410.10796](https://arxiv.org/abs/2410.10796) [find](https://scholar.google.com/scholar?q=Sachin+Goyal%2C+Christina+Baek%2C+J%2E+Zico+Kolter%2C+and+Aditi+Raghunathan%2E+2025%2E+Context%2Dparametric+inversion%3A+Why+instruction+finetuning+can+worsen+context+reliance%2E+Preprint%2C+arXiv%3A2410%2E10796%2E+%E2%80%94+%5BarXiv%3A24)
20. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, and 542 o… — [The Llama 3 Herd of Models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/)
21. Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Ming-Wei Chang. 2020. [Realm: Retrieval-augmented language model pre-training](https://md.research.elamur.ai/free/en/papers/guu-2020-realm-retrieval-augmented-language-model/). Preprint, arXiv:2002.08909.
22. Xiangkun Hu, Dongyu Ru, Lin Qiu, Qipeng Guo, Tianhang Zhang, Yang Xu, Yun Luo, Pengfei Liu, Yue Zhang, and Zheng Zhang. 2024. Refchecker: Reference-based fine-grained hallucination checker and benchmark for large language models. Preprint, arXiv:2405.14486. — [arXiv:2405.14486](https://arxiv.org/abs/2405.14486) [find](https://scholar.google.com/scholar?q=Xiangkun+Hu%2C+Dongyu+Ru%2C+Lin+Qiu%2C+Qipeng+Guo%2C+Tianhang+Zhang%2C+Yang+Xu%2C+Yun+Luo%2C+Pengfei+Liu%2C+Yue+Zhang%2C+and+Zheng+Zhang%2E+2024%2E+Refchecker%3A+Reference%2Dbased+fine%2Dgrained+hallucination+checker+and+benchma)
23. Pengcheng Huang, Zhenghao Liu, Yukun Yan, Haiyan Zhao, Xiaoyuan Yi, Hao Chen, Zhiyuan Liu, Maosong Sun, Tong Xiao, Ge Yu, and Chenyan Xiong. 2025. Parammute: Suppressing knowledge-critical ffns for faithful retrieval-augmented generation. Preprint, arXiv:2502.15543. — [arXiv:2502.15543](https://arxiv.org/abs/2502.15543) [find](https://scholar.google.com/scholar?q=Pengcheng+Huang%2C+Zhenghao+Liu%2C+Yukun+Yan%2C+Haiyan+Zhao%2C+Xiaoyuan+Yi%2C+Hao+Chen%2C+Zhiyuan+Liu%2C+Maosong+Sun%2C+Tong+Xiao%2C+Ge+Yu%2C+and+Chenyan+Xiong%2E+2025%2E+Parammute%3A+Suppressing+knowledge%2Dcritical+ffns+for+fa)
24. Jean Kaddour, Joshua Harris, Maximilian Mozes, Herbie Bradley, Roberta Raileanu, and Robert McHardy. 2023. [Challenges and applications of large language models](https://md.research.elamur.ai/free/en/papers/kaddour-2023-challenges-applications-large-language-models/). Preprint, arXiv:2307.10169.
25. Adam Tauman Kalai, Ofir Nachum, Santosh S. Vempala, and Edwin Zhang. 2025. Why language models hallucinate. Preprint, arXiv:2509.04664. — [arXiv:2509.04664](https://arxiv.org/abs/2509.04664) [find](https://scholar.google.com/scholar?q=Adam+Tauman+Kalai%2C+Ofir+Nachum%2C+Santosh+S%2E+Vempala%2C+and+Edwin+Zhang%2E+2025%2E+Why+language+models+hallucinate%2E+Preprint%2C+arXiv%3A2509%2E04664%2E+%E2%80%94+%5BarXiv%3A2509%2E04664%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2509%2E04664%29)
26. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2021. [Retrieval-augmented generation for knowledge-intensive nlp tasks](https://md.research.elamur.ai/free/en/papers/lewis-2020-retrieval-augmented-generation-knowledge-intensive/). Preprint, arXiv:2005.11401.
27. Daliang Li, Ankit Singh Rawat, Manzil Zaheer, Xin Wang, Michal Lukasik, Andreas Veit, Felix Yu, and Sanjiv Kumar. 2023a. [Large language models with controllable working memory](https://md.research.elamur.ai/free/en/papers/li-2023-large-language-models-controllable-working/). In Findings of the Association for Computational Linguistics: ACL 2023, pages 1774–1793, Toronto, Canada. Association for C…
28. Jiahang Li, Taoyu Chen, and Yuanli Wang. 2023b. Trace and edit relation associations in gpt. Preprint, arXiv:2401.02976. — [arXiv:2401.02976](https://arxiv.org/abs/2401.02976) [find](https://scholar.google.com/scholar?q=Jiahang+Li%2C+Taoyu+Chen%2C+and+Yuanli+Wang%2E+2023b%2E+Trace+and+edit+relation+associations+in+gpt%2E+Preprint%2C+arXiv%3A2401%2E02976%2E+%E2%80%94+%5BarXiv%3A2401%2E02976%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2401%2E02976%29)
29. Kenneth Li, Oam Patel, Fernanda Viégas, Hanspeter Pfister, and Martin Wattenberg. 2024a. [Inference-time intervention: Eliciting truthful answers from a language model](https://md.research.elamur.ai/free/en/papers/li-2023-inference-time-intervention-eliciting-truthful/). Preprint, arXiv:2306.03341.
30. Xiaopeng Li, Shasha Li, Shezheng Song, Jing Yang, Jun Ma, and Jie Yu. 2024b. Pmet: Precise model editing in a transformer. Preprint, arXiv:2308.08742. [find](https://scholar.google.com/scholar?q=Xiaopeng+Li%2C+Shasha+Li%2C+Shezheng+Song%2C+Jing+Yang%2C+Jun+Ma%2C+and+Jie+Yu%2E+2024b%2E+Pmet%3A+Precise+model+editing+in+a+transformer%2E+Preprint%2C+arXiv%3A2308%2E08742%2E)
31. Xinze Li, Sen Mei, Zhenghao Liu, Yukun Yan, Shuo Wang, Shi Yu, Zheni Zeng, Hao Chen, Ge Yu, Zhiyuan Liu, Maosong Sun, and Chenyan Xiong. 2025. Rag-ddr: Optimizing retrieval-augmented generation using differentiable data rewards. Preprint, arXiv:2410.13509. — [arXiv:2410.13509](https://arxiv.org/abs/2410.13509) [find](https://scholar.google.com/scholar?q=Xinze+Li%2C+Sen+Mei%2C+Zhenghao+Liu%2C+Yukun+Yan%2C+Shuo+Wang%2C+Shi+Yu%2C+Zheni+Zeng%2C+Hao+Chen%2C+Ge+Yu%2C+Zhiyuan+Liu%2C+Maosong+Sun%2C+and+Chenyan+Xiong%2E+2025%2E+Rag%2Dddr%3A+Optimizing+retrieval%2Daugmented+generation+using+)
32. Yongchao Long, Xian Wu, Yingying Zhang, Xianbin Wen, Yuxi Zhou, and Shenda Hong. 2025. Copy-paste to mitigate large language model hallucinations. Preprint, arXiv:2510.00508. — [arXiv:2510.00508](https://arxiv.org/abs/2510.00508) [find](https://scholar.google.com/scholar?q=Yongchao+Long%2C+Xian+Wu%2C+Yingying+Zhang%2C+Xianbin+Wen%2C+Yuxi+Zhou%2C+and+Shenda+Hong%2E+2025%2E+Copy%2Dpaste+to+mitigate+large+language+model+hallucinations%2E+Preprint%2C+arXiv%3A2510%2E00508%2E+%E2%80%94+%5BarXiv%3A2510%2E00508%5D%28http)
33. Varun Magesh, Faiz Surani, Matthew Dahl, Mirac Suzgun, Christopher D. Manning, and Daniel E. Ho. 2024. [Hallucination-free? assessing the reliability of leading ai legal research tools](https://md.research.elamur.ai/free/en/papers/magesh-2024-hallucination-free-assessing-reliability-leading/). Preprint, arXiv:2405.20362.
34. Piotr Matys, Jan Eliasz, Konrad Kiełczyński, Mikołaj Langner, Teddy Ferdinan, Jan Kocoń, and Przemysław Kazienko. 2025. AggTruth: Contextual Hallucination Detection Using Aggregated Attention Scores in LLMs, page 227–243. Springer Nature Switzerland. [find](https://scholar.google.com/scholar?q=Piotr+Matys%2C+Jan+Eliasz%2C+Konrad+Kie%C5%82czyn%CC%81ski%2C+Miko%C5%82aj+Langner%2C+Teddy+Ferdinan%2C+Jan+Kocon%CC%81%2C+and+Przemys%C5%82aw+Kazienko%2E+2025%2E+AggTruth%3A+Contextual+Hallucination+Detection+Using+Aggregated+Attention+Scores)
35. Kevin Meng, David Bau, Alex Andonian, and Yonatan Belinkov. 2023a. Locating and editing factual associations in gpt. Preprint, arXiv:2202.05262. [find](https://scholar.google.com/scholar?q=Kevin+Meng%2C+David+Bau%2C+Alex+Andonian%2C+and+Yonatan+Belinkov%2E+2023a%2E+Locating+and+editing+factual+associations+in+gpt%2E+Preprint%2C+arXiv%3A2202%2E05262%2E)
36. Kevin Meng, Arnab Sen Sharma, Alex Andonian, Yonatan Belinkov, and David Bau. 2023b. Mass-editing memory in a transformer. Preprint, arXiv:2210.07229. [find](https://scholar.google.com/scholar?q=Kevin+Meng%2C+Arnab+Sen+Sharma%2C+Alex+Andonian%2C+Yonatan+Belinkov%2C+and+David+Bau%2E+2023b%2E+Mass%2Dediting+memory+in+a+transformer%2E+Preprint%2C+arXiv%3A2210%2E07229%2E)
37. Abhika Mishra, Akari Asai, Vidhisha Balachandran, Yizhong Wang, Graham Neubig, Yulia Tsvetkov, and Hannaneh Hajishirzi. 2024. [Fine-grained hallucination detection and editing for language models](https://md.research.elamur.ai/free/en/papers/mishra-2024-fine-grained-hallucination-detection-editing/). Preprint, arXiv:2401.06855.
38. Eric Mitchell, Charles Lin, Antoine Bosselut, Chelsea Finn, and Christopher D. Manning. 2022. Fast model editing at scale. Preprint, arXiv:2110.11309. [find](https://scholar.google.com/scholar?q=Eric+Mitchell%2C+Charles+Lin%2C+Antoine+Bosselut%2C+Chelsea+Finn%2C+and+Christopher+D%2E+Manning%2E+2022%2E+Fast+model+editing+at+scale%2E+Preprint%2C+arXiv%3A2110%2E11309%2E)
39. Cheng Niu, Yuanhao Wu, Juno Zhu, Siliang Xu, Kashun Shum, Randy Zhong, Juntong Song, and Tong Zhang. 2024a. [Ragtruth: A hallucination corpus for developing trustworthy retrieval-augmented language models](https://md.research.elamur.ai/free/en/papers/niu-2024-ragtruth-hallucination-corpus-developing-trustworthy/). Preprint, arXiv:2401.00396.
40. Jingcheng Niu, Andrew Liu, Zining Zhu, and Gerald Penn. 2024b. What does the knowledge neuron thesis have to do with knowledge? Preprint, arXiv:2405.02421. — [arXiv:2405.02421](https://arxiv.org/abs/2405.02421) [find](https://scholar.google.com/scholar?q=Jingcheng+Niu%2C+Andrew+Liu%2C+Zining+Zhu%2C+and+Gerald+Penn%2E+2024b%2E+What+does+the+knowledge+neuron+thesis+have+to+do+with+knowledge%3F+Preprint%2C+arXiv%3A2405%2E02421%2E+%E2%80%94+%5BarXiv%3A2405%2E02421%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2)
41. 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/)
42. Keshav Santhanam, Omar Khattab, Jon Saad-Falcon, Christopher Potts, and Matei Zaharia. 2022. [Colbertv2: Effective and efficient retrieval via lightweight late interaction](https://md.research.elamur.ai/free/en/papers/santhanam-2022-colbertv2-effective-efficient-retrieval-lightweight/). Preprint, arXiv:2112.01488.
43. Shibani Santurkar, Dimitris Tsipras, Mahalaxmi Elango, David Bau, Antonio Torralba, and Aleksander Madry. 2021. Editing a classifier by rewriting its prediction rules. Preprint, arXiv:2112.01008. — [arXiv:2112.01008](https://arxiv.org/abs/2112.01008) [find](https://scholar.google.com/scholar?q=Shibani+Santurkar%2C+Dimitris+Tsipras%2C+Mahalaxmi+Elango%2C+David+Bau%2C+Antonio+Torralba%2C+and+Aleksander+Madry%2E+2021%2E+Editing+a+classifier+by+rewriting+its+prediction+rules%2E+Preprint%2C+arXiv%3A2112%2E01008%2E+%E2%80%94+%5Ba)
44. Kurt Shuster, Spencer Poff, Moya Chen, Douwe Kiela, and Jason Weston. 2021. [Retrieval augmentation reduces hallucination in conversation](https://md.research.elamur.ai/free/en/papers/shuster-2021-retrieval-augmentation-reduces-hallucination-conversation/). Preprint, arXiv:2104.07567.
45. Yi Sui, Chaozhuo Li, Chen Zhang, Dawei song, and Qiuchi Li. 2025. Bridging external and parametric knowledge: Mitigating hallucination of llms with shared-private semantic synergy in dual-stream knowledge. Preprint, arXiv:2506.06240. [find](https://scholar.google.com/scholar?q=Yi+Sui%2C+Chaozhuo+Li%2C+Chen+Zhang%2C+Dawei+song%2C+and+Qiuchi+Li%2E+2025%2E+Bridging+external+and+parametric+knowledge%3A+Mitigating+hallucination+of+llms+with+shared%2Dprivate+semantic+synergy+in+dual%2Dstream+knowl)
46. Zhongxiang Sun, Xiaoxue Zang, Kai Zheng, Yang Song, Jun Xu, Xiao Zhang, Weijie Yu, Yang Song, and Han Li. 2025. Redeep: Detecting hallucination in retrieval-augmented generation via mechanistic interpretability. Preprint, arXiv:2410.11414. — [arXiv:2410.11414](https://arxiv.org/abs/2410.11414) [find](https://scholar.google.com/scholar?q=Zhongxiang+Sun%2C+Xiaoxue+Zang%2C+Kai+Zheng%2C+Yang+Song%2C+Jun+Xu%2C+Xiao+Zhang%2C+Weijie+Yu%2C+Yang+Song%2C+and+Han+Li%2E+2025%2E+Redeep%3A+Detecting+hallucination+in+retrieval%2Daugmented+generation+via+mechanistic+interp)
47. Likun Tan, Kuan-Wei Huang, Joy Shi, and Kevin Wu. 2025. Interpdetect: Interpretable signals for detecting hallucinations in retrieval-augmented generation. Preprint, arXiv:2510.21538. — [arXiv:2510.21538](https://arxiv.org/abs/2510.21538) [find](https://scholar.google.com/scholar?q=Likun+Tan%2C+Kuan%2DWei+Huang%2C+Joy+Shi%2C+and+Kevin+Wu%2E+2025%2E+Interpdetect%3A+Interpretable+signals+for+detecting+hallucinations+in+retrieval%2Daugmented+generation%2E+Preprint%2C+arXiv%3A2510%2E21538%2E+%E2%80%94+%5BarXiv%3A2510%2E21)
48. Neeraj Varshney, Wenlin Yao, Hongming Zhang, Jianshu Chen, and Dong Yu. 2023. [A stitch in time saves nine: Detecting and mitigating hallucinations of llms by validating low-confidence generation](https://md.research.elamur.ai/free/en/papers/varshney-2023-stitch-time-saves-nine-detecting/). Preprint, arXiv:2307.03987.
49. Mengru Wang, Yunzhi Yao, Ziwen Xu, Shuofei Qiao, Shumin Deng, Peng Wang, Xiang Chen, Jia-Chen Gu, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen, and Ningyu Zhang. 2024a. Knowledge mechanisms in large language models: A survey and perspective. Preprint, arXiv:2407.15017. [find](https://scholar.google.com/scholar?q=Mengru+Wang%2C+Yunzhi+Yao%2C+Ziwen+Xu%2C+Shuofei+Qiao%2C+Shumin+Deng%2C+Peng+Wang%2C+Xiang+Chen%2C+Jia%2DChen+Gu%2C+Yong+Jiang%2C+Pengjun+Xie%2C+Fei+Huang%2C+Huajun+Chen%2C+and+Ningyu+Zhang%2E+2024a%2E+Knowledge+mechanisms+in+larg)
50. Mengru Wang, Ningyu Zhang, Ziwen Xu, Zekun Xi, Shumin Deng, Yunzhi Yao, Qishen Zhang, Linyi Yang, Jindong Wang, and Huajun Chen. 2024b. [Detoxifying large language models via knowledge editing](https://md.research.elamur.ai/free/en/papers/wang-2024-detoxifying-large-language-models-knowledge/). Preprint, arXiv:2403.14472.
51. Jason Wei, Maarten Bosma, Vincent Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M. Dai, and Quoc V Le. 2022. [Finetuned language models are zero-shot learners](https://md.research.elamur.ai/free/en/papers/wei-2021-finetuned-language-models-zero-shot/). In International Conference on Learning Representations.
52. Yang Xu, Yutai Hou, Wanxiang Che, and Min Zhang. 2023. Language anisotropic cross-lingual model editing. In Findings of the Association for Computational Linguistics: ACL 2023, page 5554–5569. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Yang+Xu%2C+Yutai+Hou%2C+Wanxiang+Che%2C+and+Min+Zhang%2E+2023%2E+Language+anisotropic+cross%2Dlingual+model+editing%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+ACL+2023%2C+page+5554%E2%80%935569%2E+Associa)
53. Ziwei Xu, Sanjay Jain, and Mohan Kankanhalli. 2025a. [Hallucination is inevitable: An innate limitation of large language models](https://md.research.elamur.ai/free/en/papers/xu-2024-hallucination-inevitable-innate-limitation-large/). Preprint, arXiv:2401.11817.
54. Ziwei Xu, Sanjay Jain, and Mohan Kankanhalli. 2025b. [Hallucination is inevitable: An innate limitation of large language models](https://md.research.elamur.ai/free/en/papers/xu-2024-hallucination-inevitable-innate-limitation-large/). Preprint, arXiv:2401.11817.
55. Yunzhi Yao, Ningyu Zhang, Zekun Xi, Mengru Wang, Ziwen Xu, Shumin Deng, and Huajun Chen. 2025. Knowledge circuits in pretrained transformers. Preprint, arXiv:2405.17969. — [arXiv:2405.17969](https://arxiv.org/abs/2405.17969) [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+2025%2E+Knowledge+circuits+in+pretrained+transformers%2E+Preprint%2C+arXiv%3A2405%2E17969%2E+%E2%80%94+%5BarXiv%3A2405%2E17969%5D%28https%3A%2F%2Fa)
56. Samuel Yeh, Sharon Li, and Tanwi Mallick. 2025. Lumina: Detecting hallucinations in rag system with context-knowledge signals. Preprint, arXiv:2509.21875. — [arXiv:2509.21875](https://arxiv.org/abs/2509.21875) [find](https://scholar.google.com/scholar?q=Samuel+Yeh%2C+Sharon+Li%2C+and+Tanwi+Mallick%2E+2025%2E+Lumina%3A+Detecting+hallucinations+in+rag+system+with+context%2Dknowledge+signals%2E+Preprint%2C+arXiv%3A2509%2E21875%2E+%E2%80%94+%5BarXiv%3A2509%2E21875%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F25)
57. Xiaowei Yuan, Zhao Yang, Yequan Wang, Shengping Liu, Jun Zhao, and Kang Liu. 2024. Discerning and resolving knowledge conflicts through adaptive decoding with contextual information-entropy constraint. In Findings of the Association for Computational Linguistics: ACL 2024, pages 3903–3922, Bangkok,… [find](https://scholar.google.com/scholar?q=Xiaowei+Yuan%2C+Zhao+Yang%2C+Yequan+Wang%2C+Shengping+Liu%2C+Jun+Zhao%2C+and+Kang+Liu%2E+2024%2E+Discerning+and+resolving+knowledge+conflicts+through+adaptive+decoding+with+contextual+information%2Dentropy+constraint)
58. Qinggang Zhang, Shengyuan Chen, Yuanchen Bei, Zheng Yuan, Huachi Zhou, Zijin Hong, Hao Chen, Yilin Xiao, Chuang Zhou, Yi Chang, and Xiao Huang. 2025. [A survey of graph retrieval-augmented generation for customized large language models](https://md.research.elamur.ai/free/en/papers/zhang-2025-survey-graph-retrieval-augmented-generation/). Preprint, arXiv:2501.13958.
59. Wenxuan Zhou, Sheng Zhang, Hoifung Poon, and Muhao Chen. 2023. [Context-faithful prompting for large language models](https://md.research.elamur.ai/free/en/papers/zhou-2023-context-faithful-prompting-large-language/). In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 14544–14556, Singapore. Association for Computational Linguistics.

*Updated 1 November 2026*