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

Multi-step Reasoning on StrategyQA

66.99Accuracy

Qwen3-8B + SFT + WeMask(TF)

45.961251.420656.8862.3394May 8, 2026May 17, 2026May 27, 2026Jun 6, 2026Jun 16, 2026Jun 26, 2026Jul 6, 2026
Updated 17d ago

Evaluation Results

MethodLinks
2026.05
66.99
2026.05
66.77
2026.05
66.69
2026.05
64.54
2026.05
64.24
2026.05
64.15
2026.05
62.62
2026.05
62.53
2026.07
61.2
2026.07
59.9
2026.07
58.8
2026.07
58.7
2026.07
58.5
2026.07
56.4
2026.07
56.2
2026.07
55.5
2026.07
53.7
2026.05
53.32
2026.07
51.7
2026.05
46.77