Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

MTRAGEval

Benchmarks

Task NameDataset NameSOTA ResultTrend
Answer GenerationMTRAGEval Subtask C 1.0 (official leaderboard)
HM58.6
4
Multi-turn RetrievalMTRAGEval Subtask A 1.0 (official leaderboard)
nDCG@557.8
3
Multi-Turn Retrieval-Augmented GenerationMTRAGEval Task B (test)
HM3 (conditional)78.27
2
Showing 3 of 3 rows