Knowledge Editing through Chain-of-Thought
Open accessAbstract
Knowledge Editing is a technique that updates large language models (LLMs) with new information to maintain their world knowledge.This approach avoids the need to rebuild the model from scratch, thereby addressing the high costs associated with frequent retraining.Among these, the in-context editing paradigm stands out for its effectiveness in integrating new knowledge while preserving the model's original capabilities.Despite its potential, existing in-context knowledge editing methods are often task-specific, focusing primarily on multi-hop QA tasks using structured knowledge triples.Moreover, their reliance on fewshot prompting for task decomposition makes them unstable and less effective in generalizing across diverse tasks.In response to these limitations, we propose EditCoT, a novel knowledge editing framework that flexibly and efficiently updates LLMs across various tasks without retraining.EditCoT works by generating a chain-of-thought (CoT) for a given input and then iteratively refining this CoT process using a CoT editor based on updated knowledge.We evaluate EditCoT across a diverse range of benchmarks, covering multiple languages and tasks.The results demonstrate that our approach achieves state-of-the-art performance while offering superior generalization, effectiveness, and stability compared to existing methods, marking a significant advancement in the field of knowledge updating 1 .
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Parameter preservation methods achieve remarkable success by adding extra modules or leveraging the in-context-learning ability of LLMs.
“In contrast, parameter preservation methods (Wang et al., 2025; Cohen et al., 2024a) achieve remarkable success by adding extra modules or leveraging the in-context-learning ability of LLMs.”
From Reason-KE++: Aligning the Process, Not Just the Outcome, for Faithful LLM Knowledge Editing (2025) · cites this paper as Wang et al., 2025
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 18
- Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei. 2020. Language models are few-shot learners. In Ad- vances in Neural Information Processing Systems 33: Annual Conference on Neural Information Process- ing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual.
- Junjie Chen, Weihang Su, Zhumin Chu, Haitao Li, Qinyao Ai, Yiqun Liu, Min Zhang, and Shaoping Ma. 2024. An automatic and cost-efficient peer-review framework for language generation evaluation. arXiv preprint arXiv:2410.12265. · arXiv
- Hengrui Gu, Kaixiong Zhou, Xiaotian Han, Ning- hao Liu, Ruobing Wang, and Xin Wang. 2024a. PokeMQA: Programmable knowledge editing for multi-hop question answering. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8069-8083, Bangkok, Thailand. Association for Computational Linguistics. · DOI
- Jia-Chen Gu, Hao-Xiang Xu, Jun-Yu Ma, Pan Lu, Zhen- Hua Ling, Kai-Wei Chang, and Nanyun Peng. 2024b. Model editing harms general abilities of large lan- guage models: Regularization to the rescue. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA, November 12-16, 2024, pages 16801-16819. Association for Computational Lin- guistics.
- Gautier Izacard, Mathilde Caron, Lucas Hosseini, Se- bastian Riedel, Piotr Bojanowski, Armand Joulin, and Edouard Grave. 2022. Unsupervised dense in- formation retrieval with contrastive learning. Trans. Mach. Learn. Res., 2022.
- Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yu- taka Matsuo, and Yusuke Iwasawa. 2022. Large lan- guage models are zero-shot reasoners. In Advances in Neural Information Processing Systems 35: An- nual Conference on Neural Information Processing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, November 28 -December 9, 2022.
- Patrick S. H. Lewis, Ethan Perez, Aleksandra Pik- tus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020. Retrieval-augmented generation for knowledge-intensive NLP tasks. In Advances in Neu- ral Information Processing Systems 33: Annual Con- ference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual.
- Kevin Meng, David Bau, Alex Andonian, and Yonatan Belinkov. 2022. Locating and editing factual associ- ations in GPT. In Advances in Neural Information Processing Systems 35: Annual Conference on Neu- ral Information Processing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, November 28 -Decem- ber 9, 2022.
- Eric Mitchell, Charles Lin, Antoine Bosselut, Christo- pher D. Manning, and Chelsea Finn. 2022. Memory- based model editing at scale. In International Con- ference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA, volume 162 of Proceedings of Machine Learning Research, pages 15817-15831.
- PMLR.
- Stephen Robertson, Hugo Zaragoza, et al. 2009. The probabilistic relevance framework: Bm25 and be- yond. Foundations and Trends® in Information Re- trieval, 3(4):333-389.
- Weihang Su, Yichen Tang, Qingyao Ai, Changyue Wang, Zhijing Wu, and Yiqun Liu. 2024b. Mitigating entity-level hallucination in large language models. arXiv preprint arXiv:2407.09417. · arXiv
- Changyue Wang, Weihang Su, Yiran Hu, Qingyao Ai, Yueyue Wu, Cheng Luo, Yiqun Liu, Min Zhang, and Shaoping Ma. 2024a. Lekube: A knowledge up- date benchmark for legal domain. In Proceedings of the 2024 Annual International ACM SIGIR Confer- ence on Research and Development in Information Retrieval in the Asia Pacific Region, pages 175-185.
- Changyue Wang, Weihang Su, Hu Yiran, Qingyao Ai, Yueyue Wu, Cheng Luo, Yiqun Liu, Min Zhang, and Shaoping Ma. 2024b. Lekube: A legal knowledge update benchmark. Peng Wang, Ningyu Zhang, Xin Xie, Yunzhi Yao, Bozhong Tian, Mengru Wang, Zekun Xi, Siyuan Cheng, Kangwei Liu, Guozhou Zheng, et al. 2023a. Easyedit: An easy-to-use knowledge editing frame- work for large language models. ArXiv preprint, abs/2308.07269. · arXiv
- Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc V. Le, Ed H. Chi, Sharan Narang, Aakanksha Chowd- hery, and Denny Zhou. 2023b. Self-consistency improves chain of thought reasoning in language models. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023. OpenReview.net.
- Zihao Wang, Anji Liu, Haowei Lin, Jiaqi Li, Xi- aojian Ma, and Yitao Liang. 2024c. Rat: Re- trieval augmented thoughts elicit context-aware reasoning in long-horizon generation. Preprint, arXiv:2403.05313. · arXiv
- Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H. Chi, Quoc V. Le, and Denny Zhou. 2022. Chain-of-thought prompting elicits reasoning in large language models. In Ad- vances in Neural Information Processing Systems 35: Annual Conference on Neural Information Process- ing Systems 2022, NeurIPS 2022, New Orleans, LA, USA, November 28 -December 9, 2022.
- An Yang, Baosong Yang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Zhou, Chengpeng Li, Chengyuan Li, Dayiheng Liu, Fei Huang, Guanting Dong, Hao- ran Wei, Huan Lin, Jialong Tang, Jialin Wang, et al. 2024. Qwen2 technical report. ArXiv preprint, abs/2407.10671. · arXiv
Cited by
Sorted by citations
Decoupling Reasoning and Knowledge Injection for In-Context Knowledge Editing
Wang, C., Su, W., Ai, Q., Zhou, Y., & Liu, Y. (2025) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2025.findings-acl.1260 · Cited by 6 · see citing works
Self-citationExDR: Explanation-driven Dynamic Retrieval Enhancement for Multimodal Fake News Detection
Ding, G., Li, Y., Zhou, Z., Lin, Z., Zha, D., & Li, J. (2026) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3805712.3809648 · Cited by 3 · see citing works
Robust Knowledge Editing via Explicit Reasoning Chains for Distractor-Resilient Multi-Hop QA
Wu, Y., Ding, L., Shen, L., & Tao, D. (2025) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2025.findings-emnlp.786 · Cited by 2 · see citing works
Dynamic and Parametric Retrieval Augmented Generation
Su, W., Dong, Q., Ai, Q., & Liu, Y. (2025) · Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region (SIGIR-AP) · doi:10.1145/3767695.3769672 · Cited by 1 · see citing works
Self-citationMitigating Context-Memory Conflicts in LLMs Through Dynamic Cognitive Reconciliation Decoding
Zhou, Y., Li, W., Lu, Y., Wang, Y., Liu, X., Wang, W., Yu, J., Zhang, M., & Li, J. (2026) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2026.3657980
Understanding the Affordances of Control in AI Reasoning for Human-AI Decision-Making
Moon, J., Huang, C., & Xiao, Z. (2026) · CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) · doi:10.1145/3772363.3798555
Learning to Edit Knowledge via Instruction-based Chain-of-Thought Prompting
Fu, J., Bai, Y., He, L., Lou, Y., Zhao, Y., Sun, L., & Su, S. (2026) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2026.acl-long.1133
Su, W., Chen, X., Wu, Y., Ai, Q., & Liu, Y. (2026) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3805712.3809928
Self-citationReason-KE++: Aligning the Process, Not Just the Outcome, for Faithful LLM Knowledge Editing
Wu, Y., Ding, L., Shen, L., & Tao, D. (2025) · arXiv · doi:10.48550/arxiv.2511.12661
- Authors
- Wang, C., Su, W., Ai, Q., Tang, Y., & Liu, Y.
- Affiliations
- Quan Cheng Laboratory (CN)
- Tsinghua University (CN)
- Venue
- Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
- Published
- 2025
- Pages
- 10684-10704
- OpenAlex
- W4416036870
- Cited by
- 3
- References
- 0
- Citations by year
- 2026: 3
- FWCI
- 2.86
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- knowledge editing, chain-of-thought, large language models, in-context learning, knowledge updating, multi-hop question answering, cross-lingual generalization, task decomposition, iterative refinement
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
