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

Multistep Reasoning on MuSR

73.33Accuracy

Spurious Universal

39.186848.050956.91565.7791Oct 21, 2025Nov 26, 2025Jan 2, 2026Feb 7, 2026Mar 16, 2026Apr 21, 2026May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.05
73.33
2026.05
71.33
2026.05
70
2026.05
70
2026.05
69.33
2026.05
67.33
2026.05
66
2026.05
64
2026.05
64
2026.05
62
2026.05
62
2026.05
62
2026.05
60.67
2026.05
60
2026.05
59.2
2026.05
58
2026.05
58
2026.05
58
2026.05
56
2026.05
56
2026.05
56
2026.05
56
2026.05
54
2026.05
54
2026.05
54
2026.05
54
2026.05
54
2026.05
54
2026.05
54
2026.05
53.7
2026.05
53.33
2026.05
52.5
2026.05
52
2026.05
52
2026.05
52
2026.05
51.6
2026.05
51.3
2026.05
50
2026.05
50
2026.05
48
2026.05
46
2026.05
46
2026.05
46
2026.05
44.2
2026.05
44.2
2026.05
43.4
2026.05
42.1
2026.05
42
2026.05
42
2025.10
41.5
2025.10
41.4
2025.10
41.4
2026.05
40.5