What External Knowledge is Preferred by LLMs? Characterizing and Exploring Chain of Evidence in Imperfect Context for Multi-Hop QA

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chang, Zhiyuan, Li, Mingyang, Jia, Xiaojun, Wang, Junjie, Huang, Yuekai, Wang, Qing, Huang, Yihao, Liu, Yang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915304451342336
author Chang, Zhiyuan
Li, Mingyang
Jia, Xiaojun
Wang, Junjie
Huang, Yuekai
Wang, Qing
Huang, Yihao
Liu, Yang
author_facet Chang, Zhiyuan
Li, Mingyang
Jia, Xiaojun
Wang, Junjie
Huang, Yuekai
Wang, Qing
Huang, Yihao
Liu, Yang
contents Incorporating external knowledge has emerged as a promising way to mitigate outdated knowledge and hallucinations in LLM. However, external knowledge is often imperfect, encompassing substantial extraneous or even inaccurate content, which interferes with the LLM's utilization of useful knowledge in the context. This paper seeks to characterize the features of preferred external knowledge and perform empirical studies in imperfect contexts. Inspired by the chain of evidence (CoE), we characterize that the knowledge preferred by LLMs should maintain both relevance to the question and mutual support among the textual pieces. Accordingly, we propose a CoE discrimination approach and conduct a comparative analysis between CoE and Non-CoE samples across significance, deceptiveness, and robustness, revealing the LLM's preference for external knowledge that aligns with CoE features. Furthermore, we selected three representative tasks (RAG-based multi-hop QA, external knowledge poisoning and poisoning defense), along with corresponding SOTA or prevalent baselines. By integrating CoE features, the variants achieved significant improvements over the original baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2412_12632
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle What External Knowledge is Preferred by LLMs? Characterizing and Exploring Chain of Evidence in Imperfect Context for Multi-Hop QA
Chang, Zhiyuan
Li, Mingyang
Jia, Xiaojun
Wang, Junjie
Huang, Yuekai
Wang, Qing
Huang, Yihao
Liu, Yang
Computation and Language
Artificial Intelligence
Incorporating external knowledge has emerged as a promising way to mitigate outdated knowledge and hallucinations in LLM. However, external knowledge is often imperfect, encompassing substantial extraneous or even inaccurate content, which interferes with the LLM's utilization of useful knowledge in the context. This paper seeks to characterize the features of preferred external knowledge and perform empirical studies in imperfect contexts. Inspired by the chain of evidence (CoE), we characterize that the knowledge preferred by LLMs should maintain both relevance to the question and mutual support among the textual pieces. Accordingly, we propose a CoE discrimination approach and conduct a comparative analysis between CoE and Non-CoE samples across significance, deceptiveness, and robustness, revealing the LLM's preference for external knowledge that aligns with CoE features. Furthermore, we selected three representative tasks (RAG-based multi-hop QA, external knowledge poisoning and poisoning defense), along with corresponding SOTA or prevalent baselines. By integrating CoE features, the variants achieved significant improvements over the original baselines.
title What External Knowledge is Preferred by LLMs? Characterizing and Exploring Chain of Evidence in Imperfect Context for Multi-Hop QA
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2412.12632