Benchmarking LLM Faithfulness in RAG with Evolving Leaderboards
This addresses the need for more reliable benchmarking of LLM hallucinations in RAG for developers and researchers, though it is incremental as it builds on existing leaderboard and detection methods.
The paper tackles the problem of measuring and benchmarking LLM faithfulness in retrieval-augmented generation (RAG) to reduce hallucinations, introducing FaithJudge, an LLM-as-a-judge framework that improves automated hallucination evaluation and an enhanced leaderboard tracking hallucination rates across tasks like summarization and question-answering.
Retrieval-augmented generation (RAG) aims to reduce hallucinations by grounding responses in external context, yet large language models (LLMs) still frequently introduce unsupported information or contradictions even when provided with relevant context. This paper presents two complementary efforts at Vectara to measure and benchmark LLM faithfulness in RAG. First, we describe our original hallucination leaderboard, which has tracked hallucination rates for LLMs since 2023 using our HHEM hallucination detection model. Motivated by limitations observed in current hallucination detection methods, we introduce FaithJudge, an LLM-as-a-judge framework that leverages a pool of diverse human-annotated hallucination examples to substantially improve the automated hallucination evaluation of LLMs. We introduce an enhanced hallucination leaderboard centered on FaithJudge that benchmarks LLMs on RAG faithfulness in summarization, question-answering, and data-to-text generation tasks. FaithJudge enables a more reliable benchmarking of LLM hallucinations in RAG and supports the development of more trustworthy generative AI systems: https://github.com/vectara/FaithJudge.