FRANQ introduces faithfulness-aware uncertainty quantification to improve detection of factual errors in RAG outputs over prior UQ methods, tested on a new long-form QA dataset annotated for both factuality and faithfulness.
In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (V olume1: Long Papers), ACL ’2024, pages 10862– 10878, Bangkok, Thailand
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Faithfulness-Aware Uncertainty Quantification for Fact-Checking the Output of Retrieval Augmented Generation
FRANQ introduces faithfulness-aware uncertainty quantification to improve detection of factual errors in RAG outputs over prior UQ methods, tested on a new long-form QA dataset annotated for both factuality and faithfulness.