ConflictBank: A Benchmark for Evaluating the Influence of Knowledge Conflicts in LLM
This addresses the issue of hallucinations in LLMs for the AI research community, though it is incremental as it builds on prior work on knowledge conflicts.
The authors tackled the problem of knowledge conflicts in large language models (LLMs) by introducing ConflictBank, a comprehensive benchmark that systematically evaluates conflicts from retrieved and encoded knowledge, resulting in the creation of over 7.4 million claim-evidence pairs and 553,000 QA pairs.
Large language models (LLMs) have achieved impressive advancements across numerous disciplines, yet the critical issue of knowledge conflicts, a major source of hallucinations, has rarely been studied. Only a few research explored the conflicts between the inherent knowledge of LLMs and the retrieved contextual knowledge. However, a thorough assessment of knowledge conflict in LLMs is still missing. Motivated by this research gap, we present ConflictBank, the first comprehensive benchmark developed to systematically evaluate knowledge conflicts from three aspects: (i) conflicts encountered in retrieved knowledge, (ii) conflicts within the models' encoded knowledge, and (iii) the interplay between these conflict forms. Our investigation delves into four model families and twelve LLM instances, meticulously analyzing conflicts stemming from misinformation, temporal discrepancies, and semantic divergences. Based on our proposed novel construction framework, we create 7,453,853 claim-evidence pairs and 553,117 QA pairs. We present numerous findings on model scale, conflict causes, and conflict types. We hope our ConflictBank benchmark will help the community better understand model behavior in conflicts and develop more reliable LLMs.