List your company

Mathematical Reasoning on TABMWP (Acc, Len)

90AccCES
77.62480.83784.0587.263May 19, 2026

Evaluation results

MethodLinks
CESBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0590460
DAPOBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0589.3397
Entropy ShapeBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0588.1387
R1-7BBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0588411
DAPOBackbone=DeepSeek-R1-Distill-1.5B2026.0579.2699
CESBackbone=DeepSeek-R1-Distill-1.5B2026.0578.1759