RAG-RewardBench: Benchmarking reward models in retrieval aug- mented generation for preference alignment
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
dataset 1