List your company

Pedagogical Dimension Evaluation on GRADE BEA 2025 (test)

77Strict F1 (MI)G3-27B GenVer
41.6450.826069.18May 27, 2026

Evaluation results

MethodLinks
G3-27B GenVerAugmentation Strategy=Gen+Verify, Model Backbone=Gemma3-27B, Authoring Group=Ours2026.0577515760
G3-12B NoAugAugmentation Strategy=NoAug, Model Backbone=Gemma3-12B, Authoring Group=Ours2026.0575505869
G3-12B GenAugmentation Strategy=Gen, Model Backbone=Gemma3-12B, Authoring Group=Ours2026.0574535558
GLM SFTEvaluation Protocol=SFT, Authoring Group=bea-jh2026.0572485954
Mathstral EnsEvaluation Protocol=Ensemble, Authoring Group=MSA2026.0572575869
GPT4o AugEvaluation Protocol=Augmentation, Authoring Group=TutorMind2026.0570———
Gemini 2.5 Pro ICLEvaluation Protocol=ICL, Authoring Group=BLCU2026.0562686756
GLM GRPOEvaluation Protocol=GRPO, Authoring Group=bea-jh2026.0556575766
DeepSeek ICLEvaluation Protocol=ICL, Authoring Group=BLCU2026.0551515447
Grok ICLEvaluation Protocol=ICL, Authoring Group=BLCU2026.0548394342
GPT4o ICLEvaluation Protocol=ICL, Authoring Group=BLCU2026.0546454944
Claude ICLEvaluation Protocol=ICL, Authoring Group=BLCU2026.0545496343
o3-mini ICLEvaluation Protocol=ICL, Authoring Group=BLCU2026.0543474945