Retrievers and rerankers built from LLMs score near random on the FACTOR factuality benchmark, far below their base models, and fail when correct answers are paraphrased.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Fact or Facsimile? Evaluating the Factual Robustness of Modern Retrievers
Retrievers and rerankers built from LLMs score near random on the FACTOR factuality benchmark, far below their base models, and fail when correct answers are paraphrased.