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

General Reasoning on Aggregated Reasoning Tasks 10-Task Combination

52.2Average Score

Qwen2.5-7B + E3-TIR (Ours)

22.76830.40938.0545.691Apr 10, 2026
Updated 3mo ago

Evaluation Results

MethodLinks
2026.04
52.27.5
2026.04
49.97.7
2026.04
497.9
2026.04
47.87.4
2026.04
46.76.2
2026.04
45.77.1
2026.04
44.97.7
2026.04
44.26.1
2026.04
43.27.9
2026.04
39.28.2
2026.04
34.88.4
2026.04
32.97.4
2026.04
30.7-
2026.04
26.8-
2026.04
23.9-