Skip to content
conftrace_

internal prior

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

external evidence (1) conflicting information (1) token probability (1) large language model (1) retrieval-augmented generation (1) hallucination mitigation (1) benchmark dataset (1) information retrieval (1) question answering (1)

Papers

ClashEval: Quantifying the tug-of-war between an LLM’s internal prior and external evidence NeurIPS 2024