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

Fact Verification on M2RAG

70Accuracy

Ours-top3

37.7646.1354.562.87Apr 27, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
7073.75
2026.04
6869.8
2026.04
6667.43
2026.04
6364.31
2026.04
6365.96
2026.04
6263.65
2026.04
6262.95
2026.04
6263.74
2026.04
6062.44
2026.04
6062.26
2026.04
5961.15
2026.04
5841.13
2026.04
5859.23
2026.04
5656.63
2026.04
5538.25
2026.04
5555.04
2026.04
5234.93
2026.04
5251.51
2026.04
5252.02
2026.04
5150.67
2026.04
5033.18
2026.04
5034.26
2026.04
4731.14
2026.04
4746.89
2026.04
4544.81
2026.04
4343.07
2026.04
4138.96
2026.04
3935.76