Benchmark
MultiDoc2Dial
A multi-document dialogue dataset used to evaluate the performance of retrieval and question-answering systems in multi-document scenarios.
Timeline (last 90 days)
Oct 27
On the MultiDoc2Dial dataset, DecisionRAG, without task-specific training and without a first-stage retriever, achieves 13.8 higher R@1 points than dense retrieval using an 8B reranker.
Unverified50%