List your company

Mathematical Reasoning on Robustness Evaluation Suite (GSMIR, MMLU-P, OBook-P)

7.39GSMIR ScoreInT
2.78283.97895.1756.3711May 12, 2026

Evaluation results

MethodLinks
InT2026.057.398.2517.9211.19
SAGE2026.054.347.5419.6210.5
Scaf-GRPO2026.053.579.1219.0310.57
Seirênes2026.053.054.9116.218.06
Qwen3-4B-Instruct2026.052.969.3919.0310.46