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

Multi-step Soft Reasoning on MuSR

65.3Accuracy

QwQ-32B

24.2234.88545.5556.215Jun 2, 2026Jun 7, 2026Jun 13, 2026Jun 19, 2026Jun 24, 2026Jun 30, 2026Jul 6, 2026
Updated 18d ago

Evaluation Results

MethodLinks
2026.06
65.354.173.666.580.671.377.167.2
2026.06
63.966.6726861.276.77964.8
2026.07
59.3-------
2026.07
56.8-------
2026.07
55.2-------
2026.07
52.5-------
2026.07
50.9-------
2026.07
29.5-------
2026.07
29.3-------
2026.07
28.2-------
2026.07
27.8-------
2026.07
25.8-------