Knowledge Conflicts for LLMs: A Survey
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917701199331328 |
|---|---|
| author | Xu, Rongwu Qi, Zehan Guo, Zhijiang Wang, Cunxiang Wang, Hongru Zhang, Yue Xu, Wei |
| author_facet | Xu, Rongwu Qi, Zehan Guo, Zhijiang Wang, Cunxiang Wang, Hongru Zhang, Yue Xu, Wei |
| contents | This survey provides an in-depth analysis of knowledge conflicts for large language models (LLMs), highlighting the complex challenges they encounter when blending contextual and parametric knowledge. Our focus is on three categories of knowledge conflicts: context-memory, inter-context, and intra-memory conflict. These conflicts can significantly impact the trustworthiness and performance of LLMs, especially in real-world applications where noise and misinformation are common. By categorizing these conflicts, exploring the causes, examining the behaviors of LLMs under such conflicts, and reviewing available solutions, this survey aims to shed light on strategies for improving the robustness of LLMs, thereby serving as a valuable resource for advancing research in this evolving area. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_08319 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Knowledge Conflicts for LLMs: A Survey Xu, Rongwu Qi, Zehan Guo, Zhijiang Wang, Cunxiang Wang, Hongru Zhang, Yue Xu, Wei Computation and Language Artificial Intelligence Information Retrieval Machine Learning This survey provides an in-depth analysis of knowledge conflicts for large language models (LLMs), highlighting the complex challenges they encounter when blending contextual and parametric knowledge. Our focus is on three categories of knowledge conflicts: context-memory, inter-context, and intra-memory conflict. These conflicts can significantly impact the trustworthiness and performance of LLMs, especially in real-world applications where noise and misinformation are common. By categorizing these conflicts, exploring the causes, examining the behaviors of LLMs under such conflicts, and reviewing available solutions, this survey aims to shed light on strategies for improving the robustness of LLMs, thereby serving as a valuable resource for advancing research in this evolving area. |
| title | Knowledge Conflicts for LLMs: A Survey |
| topic | Computation and Language Artificial Intelligence Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2403.08319 |