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

Instruction Following on RecRM-Bench

72.66Accuracy

Ours

24.830437.247749.66562.0823May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.05
72.6672.4
2026.05
64.8465.19
2026.05
55.4758.33
2026.05
43.7548.16
2026.05
36.840.33
2026.05
35.2939.5
2026.05
30.4729.94
2026.05
26.6726.64