Breaking the Reference Bottleneck via Learning to Rewrite Conversational Queries without Gold Reference Passages
A novel reference-free preference optimization framework D UAL R EFORM that generates pseudo reference passages from commonly-encountered conversational datasets containing only queries and responses that achieves 96.9–99.1% of the retrieval accuracy attainable only with reference passages.
Doyoung Kim, Youngjun Lee, Joeun Kim et al.
· 0 citations