List your company
ResearchFormerly WizWand

Scientific and General Reasoning on Theorem QA

55.75Pass@1MulFeRL
17.5327.452537.37547.2975Jun 3, 2025Jul 13, 2025Aug 22, 2025Oct 1, 2025Nov 10, 2025Dec 20, 2025Jan 30, 2026
Updated 8mo ago

Evaluation results

MethodLinks
MulFeRLBackbone=Qwen3-4B-Inst, Training Strategy=Multi-turn Feedback-guided Reinforcement Learning2026.0155.75
Critique-GRPOTraining Data Volume=4k, Critique Mode=CoT-Critique2025.0651.4
Critique-GRPOBackbone=Qwen3-4B-Inst, Training Strategy=Reinforcement Learning-based Finetuning2026.0151.05
Dr.GRPOBackbone=Qwen3-4B-Inst, Training Strategy=Reinforcement Learning-based Finetuning2026.0150.68
GRPOBackbone=Qwen3-4B-Inst, Training Strategy=Reinforcement Learning-based Finetuning2026.0149.85
MulFeRLBackbone=Qwen2.5-7B-Base, Training Strategy=Multi-turn Feedback-guided Reinforcement Learning2026.0143.08
CITL-FTBackbone=Qwen3-4B-Inst, Training Strategy=Supervised Learning-based Finetuning2026.0142.65
SFTBackbone=Qwen3-4B-Inst, Training Strategy=Supervised Learning-based Finetuning2026.0141.93
RAFTBackbone=Qwen3-4B-Inst, Training Strategy=Supervised Learning-based Finetuning2026.0141.05
Dr.GRPOBackbone=Qwen2.5-7B-Base, Training Strategy=Reinforcement Learning-based Finetuning2026.0140.08
Qwen3-4B-InstBackbone=Qwen3-4B-Inst, Training Strategy=Base Model2026.0139.85
Critique-GRPOBackbone=Qwen2.5-7B-Base, Training Strategy=Reinforcement Learning-based Finetuning2026.0139.75
GRPOBackbone=Qwen2.5-7B-Base, Training Strategy=Reinforcement Learning-based Finetuning2026.0137.55
Qwen2.5-Math-7B-BaseTraining Data Volume=None2025.0626.4
SFTBackbone=Qwen2.5-7B-Base, Training Strategy=Supervised Learning-based Finetuning2026.0124.1
CITL-FTBackbone=Qwen2.5-7B-Base, Training Strategy=Supervised Learning-based Finetuning2026.0123.65
RAFTBackbone=Qwen2.5-7B-Base, Training Strategy=Supervised Learning-based Finetuning2026.0121.9
Qwen2.5-7B-BaseBackbone=Qwen2.5-7B-Base, Training Strategy=Base Model2026.0119