The DOI matches its Crossref record.
InstructEdit: Instruction-Based Knowledge Editing for Large Language Models
Open accessAbstract
Knowledge editing for large language models can offer an efficient solution to alter a model’s behavior without negatively impacting the overall performance. However, the current approaches encounter issues with limited generalizability across tasks, necessitating one distinct editor for each task, significantly hindering the broader applications. To address this, we take the first step to analyze the multi-task generalization issue in knowledge editing. Specifically, we develop an instruction-based editing technique, termed InstructEdit, which facilitates the editor's adaptation to various task performances simultaneously using simple instructions. With only one unified editor for each LLM, we empirically demonstrate that InstructEdit can improve the editor's control, leading to an average 14.86% increase in Reliability in multi-task editing setting. Furthermore, experiments involving holdout unseen task illustrate that InstructEdit consistently surpass previous strong baselines. To further investigate the underlying mechanisms of instruction-based knowledge editing, we analyze the principal components of the editing gradient directions, which unveils that instructions can help control optimization direction with stronger OOD generalization.
Overview
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
Verification
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?
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“Datasets such as Counterfact (Meng et al., 2022), MQuAKE (Zhong et al., 2023a), KnowEdit (Zhang et al., 2024), and QAEdit (Yang et al., 2025) are excluded from our evaluation, due to their limited scale to support meaningful large-scale lifelong model editing.”
From UltraEdit: Training-, Subject-, and Memory-Free Lifelong Editing in Language Models (2025) · cites this paper as Zhang et al., 2024
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
No references could be resolved yet.
Research works printed but not linked: 38
- Akyürek et al., 2023] Afra Feyza Akyürek, Eric Pan, Garry Kuwanto, and Derry Wijaya. Dune: Dataset for unified editing. In EMNLP, 2023.
- Nicola De Cao, Wilker Aziz, and Ivan Titov. Editing factual knowledge in language models. In EMNLP, 2021.
- Siyuan Cheng, Bozhong Tian, Qingbin Liu, Xi Chen, Yongheng Wang, Huajun Chen, and Ningyu Zhang. Can we edit multimodal large language models? In EMNLP, 2023.
- Siyuan Cheng, Ningyu Zhang, Bozhong Tian, Xi Chen, Qingbing Liu, and Huajun Chen. Editing language model-based knowledge graph embeddings. In AAAI, 2024.
- Roi Cohen, Eden Biran, Ori Yoran, Amir Globerson, and Mor Geva. Evaluating the ripple ef- fects of knowledge editing in language models. CoRR, abs/2307.12976, 2023. · arXiv
- Damai Dai, Li Dong, Yaru Hao, Zhifang Sui, Baobao Chang, and Furu Wei. Knowledge neurons in pretrained transformers. In ACL, 2022.
- Dong et al., 2022] Qingxiu Dong, Damai Dai, Yifan Song, Jingjing Xu, Zhifang Sui, and Lei Li. Calibrating factual knowledge in pretrained language models. In EMNLP, 2022.
- Rohit Gandikota, Joanna Materzyn- ska, Jaden Fiotto-Kaufman, and David Bau. Erasing con- cepts from diffusion models. CoRR, abs/2303.07345, 2023. · arXiv
- Gupta et al., 2023] Anshita Gupta, Debanjan Mondal, Ak- shay Krishna Sheshadri, Wenlong Zhao, Xiang Lorraine Li, Sarah Wiegreffe, and Niket Tandon. Editing commonsense knowledge in GPT. In EMNLP, 2023.
- Hartvigsen et al., 2023] Tom Hartvigsen, Swami Sankara- narayanan, Hamid Palangi, Yoon Kim, and Marzyeh Ghas- semi. Aging with GRACE: lifelong model editing with discrete key-value adaptors. In NeurIPS, 2023.
- Huang et al., 2023] Zeyu Huang, Yikang Shen, Xiaofeng Zhang, Jie Zhou, Wenge Rong, and Zhang Xiong. Transformer-patcher: One mistake worth one neuron. In ICLR, 2023.
- Gabriel Ilharco, Marco Tulio Ribeiro, Mitchell Wortsman, Ludwig Schmidt, Hannaneh Hajishirzi, and Ali Farhadi. Editing models with task arithmetic. In ICLR, 2023.
- Li et al., 2023a] Kenneth Li, Oam Patel, Fernanda Viégas, Hanspeter Pfister, and Martin Wattenberg. Inference-time intervention: Eliciting truthful answers from a language model. In NeurIPS, 2023.
- Li et al., 2023c] Zhoubo Li, Ningyu Zhang, Yunzhi Yao, Mengru Wang, Xi Chen, and Huajun Chen. Unveiling the pitfalls of knowledge editing for large language models. arXiv preprint arXiv:2310.02129, 2023. · arXiv
- Li et al., 2023d] Zichao Li, Ines Arous, Siva Reddy, and Jackie C. K. Cheung. Evaluating dependencies in fact editing for language models: Specificity and implication awareness. In EMNLP, 2023.
- Liang et al., 2022] Xiaozhuan Liang, Ningyu Zhang, Siyuan Cheng, Zhenru Zhang, Chuanqi Tan, and Huajun Chen. Contrastive demonstration tuning for pre-trained language models. In EMNLP, 2022.
- Meng et al., 2022a] Kevin Meng, David Bau, Alex Ando- nian, and Yonatan Belinkov. Locating and editing factual knowledge in GPT. In NeurIPS, 2022.
- Meng et al., 2022b] Kevin Meng, Arnab Sen Sharma, Alex Andonian, Yonatan Belinkov, and David Bau. Mass-editing memory in a transformer. CoRR, abs/2210.07229, 2022. · arXiv
- Mitchell et al., 2022a] Eric Mitchell, Charles Lin, Antoine Bosselut, Chelsea Finn, and Christopher D. Manning. Fast model editing at scale. In ICLR, 2022.
- Mitchell et al., 2022b] Eric Mitchell, Charles Lin, Antoine Bosselut, Christopher D. Manning, and Chelsea Finn. Memory-based model editing at scale. In ICML, 2022.
- Yasumasa Onoe, Michael J. Q. Zhang, Shankar Padmanabhan, Greg Durrett, and Eunsol Choi. Can lms learn new entities from descriptions? challenges in propagating injected knowledge. In ACL, 2023. [OpenAI, 2022] OpenAI. Introducing chatgpt. https://openai. com/blog/chatgpt, 2022. [OpenAI, 2023] OpenAI. GPT-4 technical report. CoRR, abs/2303.08774, 2023. · arXiv
- Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul F. Christiano, Jan Leike, and Ryan Lowe. Training language models to follow instructions with human feedback. In NeurIPS, 2022.
- Pinter and Elhadad, 2023] Yuval Pinter and Michael El- hadad. Emptying the ocean with a spoon: Should we edit models? In EMNLP, 2023.
- Anton Sinitsin, Vsevolod Plokhotnyuk, Dmitry V. Pyrkin, Sergei Popov, and Artem Babenko. Ed- itable neural networks. In ICLR, 2020.
- Su et al., 2023] Hongjin Su, Weijia Shi, Jungo Kasai, Yizhong Wang, Yushi Hu, Mari Ostendorf, Wen-tau Yih, Noah A. Smith, Luke Zettlemoyer, and Tao Yu. One em- bedder, any task: Instruction-finetuned text embeddings. In ACL, 2023.
- Tan et al., 2024] Chenmien Tan, Ge Zhang, and Jie Fu. Mas- sive editing for large language models via meta learning. In ICLR, 2024.
- Touvron et al., 2023] Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurélien Rodriguez, Armand Joulin, Edouard Grave, and Guillaume Lample. Llama: Open and efficient foundation language models. CoRR, abs/2302.13971, 2023. · arXiv
- Wang et al., 2023b] Peng Wang, Ningyu Zhang, Bozhong Tian, Zekun Xi, Yunzhi Yao, Ziwen Xu, Mengru Wang, Shengyu Mao, Xiaohan Wang, Siyuan Cheng, Kangwei Liu, Yuansheng Ni, Guozhou Zheng, and Huajun Chen. Easyedit: An easy-to-use knowledge editing framework for large language models. CoRR, abs/2308.07269, 2023. · arXiv
- Jason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M. Dai, and Quoc V. Le. Finetuned language models are zero-shot learners. In ICLR, 2022.
- Suhang Wu, Minlong Peng, Yue Chen, Jin- song Su, and Mingming Sun. Eva-kellm: A new bench- mark for evaluating knowledge editing of llms. CoRR, abs/2308.09954, 2023. · arXiv
- Yang Xu, Yutai Hou, and Wanxiang Che. Language anisotropic cross-lingual model editing. ArXiv, abs/2205.12677, 2022. · arXiv
- Yunzhi Yao, Peng Wang, Bozhong Tian, Siyuan Cheng, Zhoubo Li, Shumin Deng, Huajun Chen, and Ningyu Zhang. Editing large language models: Problems, methods, and opportunities. In EMNLP, 2023.
- Xunjian Yin, Jin Jiang, Liming Yang, and Xiaojun Wan. History matters: Temporal knowledge editing in large language model. In AAAI, 2024.
- Lang Yu, Qin Chen, Jie Zhou, and Liang He. MELO: enhancing model editing with neuron-indexed dynamic lora. In AAAI, 2024.
- Zhang et al., 2023a] Ningyu Zhang, Yunzhi Yao, and Shumin Deng. Editing large language models. In AACL: Tutorial Abstract, 2023.
- Zhang et al., 2023b] Shengyu Zhang, Linfeng Dong, Xiaoya Li, Sen Zhang, Xiaofei Sun, Shuhe Wang, Jiwei Li, Runyi Hu, Tianwei Zhang, Fei Wu, and Guoyin Wang. Instruc- tion tuning for large language models: A survey. CoRR, abs/2308.10792, 2023. · arXiv
- Ce Zheng, Lei Li, Qingxiu Dong, Yuxuan Fan, Zhiyong Wu, Jingjing Xu, and Baobao Chang. Can we edit factual knowledge by in-context learning? In EMNLP, 2023.
- Zhong et al., 2023] Zexuan Zhong, Zhengxuan Wu, Christo- pher D. Manning, Christopher Potts, and Danqi Chen. Mquake: Assessing knowledge editing in language models via multi-hop questions. In EMNLP, 2023.
Cited by
Sorted by citations
UltraEdit: Training-, Subject-, and Memory-Free Lifelong Editing in Language Models
Gu, X., Huang, Z., Gu, J.-C., & Zhang, K. (2025) · arXiv · doi:10.48550/arxiv.2505.14679 · Cited by 10
Metadata
- Authors
- Zhang, N., Tian, B., Cheng, S., Liang, X., Hu, Y., Xue, K., Gou, Y., Chen, X., & Chen, H.
- Affiliations
- Zhejiang University (CN)
- Tencent (China) (CN)
- Zhejiang University
- Venue
- Proceedings of the Thirty-ThirdInternational Joint Conference on Artificial Intelligence
- Published
- 2024-08
- Pages
- 6633-6641
- OpenAlex
- W4401023625
- Cited by
- 3
- References
- 0
- Citations by year
- 2026: 1 · 2025: 1 · 2024: 1
- FWCI
- 0.89
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- International Joint Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Tencent
- National Natural Science Foundation of China (226-2023-00138, 62206246, 2023a)
- Zhejiang University
- Fundamental Research Funds for the Central Universities (226-2023-00138)
- Natural Science Foundation of Zhejiang Province (LGG22F030011)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- knowledge editing, large language models, generalization to unseen tasks
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
