A self-selection RAG method, trained with DPO on one-correct-one-wrong answer pairs, improves accuracy on NQ and TriviaQA in most settings.
Christiano, Jan Leike, Tom B
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Optimizing Knowledge Integration in Retrieval-Augmented Generation with Self-Selection
A self-selection RAG method, trained with DPO on one-correct-one-wrong answer pairs, improves accuracy on NQ and TriviaQA in most settings.