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

Factual Consistency on RecRM-Bench

70.71Accuracy (%)

Ours

40.23848.14956.0663.971May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.05
70.7182.84
2026.05
67.3480.48
2026.05
64.9878.78
2026.05
62.9677.27
2026.05
56.5772.26
2026.05
53.269.45
2026.05
43.4360.56
2026.05
41.4158.57