Enhancing Multi-hop Reasoning through Knowledge Erasure in Large Language Model Editing
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916366248837120 |
|---|---|
| author | Zhang, Mengqi Fang, Bowen Liu, Qiang Ren, Pengjie Wu, Shu Chen, Zhumin Wang, Liang |
| author_facet | Zhang, Mengqi Fang, Bowen Liu, Qiang Ren, Pengjie Wu, Shu Chen, Zhumin Wang, Liang |
| contents | Large language models (LLMs) face challenges with internal knowledge inaccuracies and outdated information. Knowledge editing has emerged as a pivotal approach to mitigate these issues. Although current knowledge editing techniques exhibit promising performance in single-hop reasoning tasks, they show limitations when applied to multi-hop reasoning. Drawing on cognitive neuroscience and the operational mechanisms of LLMs, we hypothesize that the residual single-hop knowledge after editing causes edited models to revert to their original answers when processing multi-hop questions, thereby undermining their performance in multihop reasoning tasks. To validate this hypothesis, we conduct a series of experiments that empirically confirm our assumptions. Building on the validated hypothesis, we propose a novel knowledge editing method that incorporates a Knowledge Erasure mechanism for Large language model Editing (KELE). Specifically, we design an erasure function for residual knowledge and an injection function for new knowledge. Through joint optimization, we derive the optimal recall vector, which is subsequently utilized within a rank-one editing framework to update the parameters of targeted model layers. Extensive experiments on GPT-J and GPT-2 XL demonstrate that KELE substantially enhances the multi-hop reasoning capability of edited LLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_12456 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing Multi-hop Reasoning through Knowledge Erasure in Large Language Model Editing Zhang, Mengqi Fang, Bowen Liu, Qiang Ren, Pengjie Wu, Shu Chen, Zhumin Wang, Liang Computation and Language Large language models (LLMs) face challenges with internal knowledge inaccuracies and outdated information. Knowledge editing has emerged as a pivotal approach to mitigate these issues. Although current knowledge editing techniques exhibit promising performance in single-hop reasoning tasks, they show limitations when applied to multi-hop reasoning. Drawing on cognitive neuroscience and the operational mechanisms of LLMs, we hypothesize that the residual single-hop knowledge after editing causes edited models to revert to their original answers when processing multi-hop questions, thereby undermining their performance in multihop reasoning tasks. To validate this hypothesis, we conduct a series of experiments that empirically confirm our assumptions. Building on the validated hypothesis, we propose a novel knowledge editing method that incorporates a Knowledge Erasure mechanism for Large language model Editing (KELE). Specifically, we design an erasure function for residual knowledge and an injection function for new knowledge. Through joint optimization, we derive the optimal recall vector, which is subsequently utilized within a rank-one editing framework to update the parameters of targeted model layers. Extensive experiments on GPT-J and GPT-2 XL demonstrate that KELE substantially enhances the multi-hop reasoning capability of edited LLMs. |
| title | Enhancing Multi-hop Reasoning through Knowledge Erasure in Large Language Model Editing |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2408.12456 |