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

Judge alignment prediction on Arena Hard English v2.0

55.78Coding Accuracy

TRACE

43.757646.87885053.1212May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.05
55.785064.37
2026.05
52.9653.4558.62
2026.05
49.2147.7950.57
44.2247.8536.78