InstructEdit: Instruction-based Knowledge Editing for Large Language Models

• URL: http://arxiv.org/abs/2402.16123v2
• Date: Sun, 28 Apr 2024 12:03:38 GMT
• Title: InstructEdit: Instruction-based Knowledge Editing for Large Language Models
• Authors: Ningyu Zhang, Bozhong Tian, Siyuan Cheng, Xiaozhuan Liang, Yi Hu, Kouying Xue, Yanjie Gou, Xi Chen, Huajun Chen,
• Abstract summary: We develop an instruction-based editing technique, termed InstructEdit, which facilitates the editor's adaptation to various task performances simultaneously using simple instructions. Experiments involving holdout unseen task illustrate that InstructEdit consistently surpass previous strong baselines.
• Score: 39.2147118489123
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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. Code and datasets are available in https://github.com/zjunlp/EasyEdit.

Related papers

• Uncovering Overfitting in Large Language Model Editing [35.55260822503773]
  We identify and investigate the phenomenon of Editing Overfit, where edited models assign disproportionately high probabilities to the edit target. We propose a new plug-and-play strategy called Learn to Inference (LTI), which introduce a Multi-stage Inference Constraint module to guide the edited models in recalling new knowledge.
  arXiv  Detail & Related papers  (2024-10-10T11:09:00Z)
• LEMoE: Advanced Mixture of Experts Adaptor for Lifelong Model Editing of Large Language Models [30.831866499812925]
  Large language models (LLMs) require continual knowledge updates to stay abreast of the ever-changing world facts. We introduce LEMoE, an advanced Mixture of Experts (MoE) adaptor for lifelong model editing.
  arXiv  Detail & Related papers  (2024-06-28T16:17:41Z)
• InstructBrush: Learning Attention-based Instruction Optimization for Image Editing [54.07526261513434]
  InstructBrush is an inversion method for instruction-based image editing methods. It extracts editing effects from image pairs as editing instructions, which are further applied for image editing. Our approach achieves superior performance in editing and is more semantically consistent with the target editing effects.
  arXiv  Detail & Related papers  (2024-03-27T15:03:38Z)
• Knowledge Graph Enhanced Large Language Model Editing [37.6721061644483]
  Large language models (LLMs) are pivotal in advancing natural language processing (NLP) tasks. Existing editing methods struggle to track and incorporate changes in knowledge associated with edits. We propose a novel model editing method that leverages knowledge graphs for enhancing LLM editing, namely GLAME.
  arXiv  Detail & Related papers  (2024-02-21T07:52:26Z)
• The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse [58.0132400208411]
  Even a single edit can trigger model collapse, manifesting as significant performance degradation in various benchmark tasks. benchmarking Large Language Models after each edit is impractically time-consuming and resource-intensive. We have utilized GPT-3.5 to develop a new dataset, HardEdit, based on hard cases.
  arXiv  Detail & Related papers  (2024-02-15T01:50:38Z)
• SmartEdit: Exploring Complex Instruction-based Image Editing with Multimodal Large Language Models [91.22477798288003]
  This paper introduces SmartEdit, a novel approach to instruction-based image editing. It exploits Multimodal Large Language Models (MLLMs) to enhance their understanding and reasoning capabilities. We show that a small amount of complex instruction editing data can effectively stimulate SmartEdit's editing capabilities for more complex instructions.
  arXiv  Detail & Related papers  (2023-12-11T17:54:11Z)
• Emu Edit: Precise Image Editing via Recognition and Generation Tasks [62.95717180730946]
  We present Emu Edit, a multi-task image editing model which sets state-of-the-art results in instruction-based image editing. We train it to multi-task across an unprecedented range of tasks, such as region-based editing, free-form editing, and Computer Vision tasks. We show that Emu Edit can generalize to new tasks, such as image inpainting, super-resolution, and compositions of editing tasks, with just a few labeled examples.
  arXiv  Detail & Related papers  (2023-11-16T18:55:58Z)
• XATU: A Fine-grained Instruction-based Benchmark for Explainable Text Updates [7.660511135287692]
  This paper introduces XATU, the first benchmark specifically designed for fine-grained instruction-based explainable text editing. XATU considers finer-grained text editing tasks of varying difficulty, incorporating lexical, syntactic, semantic, and knowledge-intensive edit aspects. We demonstrate the effectiveness of instruction tuning and the impact of underlying architecture across various editing tasks.
  arXiv  Detail & Related papers  (2023-09-20T04:58:59Z)
• CoEdIT: Text Editing by Task-Specific Instruction Tuning [18.824571167583432]
  CoEdIT is a state-of-the-art text editing system for writing assistance. It takes instructions from the user specifying the attributes of the desired text, and outputs the edited text. We present a large language model fine-tuned on a diverse collection of task-specific instructions for text editing.
  arXiv  Detail & Related papers  (2023-05-17T00:05:24Z)
• Learning Structural Edits via Incremental Tree Transformations [102.64394890816178]
  We present a generic model for incremental editing of structured data (i.e., "structural edits") Our editor learns to iteratively generate tree edits (e.g., deleting or adding a subtree) and applies them to the partially edited data. We evaluate our proposed editor on two source code edit datasets, where results show that, with the proposed edit encoder, our editor significantly improves accuracy over previous approaches.
  arXiv  Detail & Related papers  (2021-01-28T16:11:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.