Propagation and Pitfalls: Reasoning-based Assessment of Knowledge Editing through Counterfactual Tasks
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929230141456384 |
|---|---|
| author | Hua, Wenyue Guo, Jiang Dong, Mingwen Zhu, Henghui Ng, Patrick Wang, Zhiguo |
| author_facet | Hua, Wenyue Guo, Jiang Dong, Mingwen Zhu, Henghui Ng, Patrick Wang, Zhiguo |
| contents | Current approaches of knowledge editing struggle to effectively propagate updates to interconnected facts. In this work, we delve into the barriers that hinder the appropriate propagation of updated knowledge within these models for accurate reasoning. To support our analysis, we introduce a novel reasoning-based benchmark -- ReCoE (Reasoning-based Counterfactual Editing dataset) -- which covers six common reasoning schemes in real world. We conduct a thorough analysis of existing knowledge editing techniques, including input augmentation, finetuning, and locate-and-edit. We found that all model editing methods show notably low performance on this dataset, especially in certain reasoning schemes. Our analysis over the chain-of-thought generation of edited models further uncover key reasons behind the inadequacy of existing knowledge editing methods from a reasoning standpoint, involving aspects on fact-wise editing, fact recall ability, and coherence in generation. We will make our benchmark publicly available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_17585 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Propagation and Pitfalls: Reasoning-based Assessment of Knowledge Editing through Counterfactual Tasks Hua, Wenyue Guo, Jiang Dong, Mingwen Zhu, Henghui Ng, Patrick Wang, Zhiguo Computation and Language Artificial Intelligence Machine Learning Methodology Current approaches of knowledge editing struggle to effectively propagate updates to interconnected facts. In this work, we delve into the barriers that hinder the appropriate propagation of updated knowledge within these models for accurate reasoning. To support our analysis, we introduce a novel reasoning-based benchmark -- ReCoE (Reasoning-based Counterfactual Editing dataset) -- which covers six common reasoning schemes in real world. We conduct a thorough analysis of existing knowledge editing techniques, including input augmentation, finetuning, and locate-and-edit. We found that all model editing methods show notably low performance on this dataset, especially in certain reasoning schemes. Our analysis over the chain-of-thought generation of edited models further uncover key reasons behind the inadequacy of existing knowledge editing methods from a reasoning standpoint, involving aspects on fact-wise editing, fact recall ability, and coherence in generation. We will make our benchmark publicly available. |
| title | Propagation and Pitfalls: Reasoning-based Assessment of Knowledge Editing through Counterfactual Tasks |
| topic | Computation and Language Artificial Intelligence Machine Learning Methodology |
| url | https://arxiv.org/abs/2401.17585 |