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

Mathematical Reasoning on Olympiad

70.9Accuracy

DeepSeek-R1-Distill-Qwen-7B + ERC-DAPO

31.69241.87152.0562.229Jun 30, 2025Aug 13, 2025Sep 26, 2025Nov 10, 2025Dec 24, 2025Feb 6, 2026Mar 23, 2026
Updated 25d ago

Evaluation Results

MethodLinks
2025.12
70.9
2025.12
69.9
2025.12
67
2025.12
65.6
2025.12
61
2026.01
59.05
2025.12
58.6
2026.03
58.22
2026.03
58.22
2026.01
58
2025.06
57.9
2026.02
57.2
2026.01
56.92
2025.06
56.9
2025.12
56.2
2026.03
55.56
2026.03
55.56
2026.03
55.41
2026.02
54.1
2026.02
54.1
2026.03
53.78
2026.03
53.33
2025.06
52.4
2026.01
52.2
2025.12
51.8
2026.02
51.7
2026.01
51.61
2026.03
51.11
2026.02
50.8
2026.02
50.8
2026.02
50.7
2026.02
50.7
2026.02
50.5
2026.02
50.5
2026.02
50.4
2026.01
49.99
2026.02
49.9
2026.02
49.9
2026.02
49.9
2026.02
49.6
2025.06
49.6
2026.03
49.33
2026.03
49.04
2026.02
48.9
2026.01
48.32
2025.12
48.15
2025.12
47.8
2025.12
47.7
2026.02
47.4
2026.02
47.1
2025.12
46.81
2026.03
46.81
2026.03
46.67
2026.02
46.5
2025.12
46.5
2025.12
46.4
2025.12
46.3
2026.03
46.22
2026.01
46.14
2026.01
45.97
2026.03
45.93
2026.03
45.93
2025.06
45.9
2025.12
45.62
2026.02
45.5
45.33
2025.12
45.33
2026.02
45.3
2025.12
44.74
2025.12
44.59
2026.03
44.3
2026.03
44.3
2025.12
44.15
2025.12
44
2026.03
44
2026.03
43.85
2026.01
43.76
2026.03
43.56
2025.12
43.4
2026.02
43
2025.12
42.8
2025.12
42.7
2025.12
42.5
2025.12
42.4
2025.06
41.8
2025.12
41.5
2025.12
40.6
2025.12
40.4
2025.12
40.15
2025.06
38.4
2025.06
37.6
2025.06
36.7
2026.03
36.3
2025.12
35.7
2025.12
34.8
2026.03
33.8
2025.06
33.7
2025.06
33.5
2025.06
33.3
33.2
Showing 100 of 137 rows