List your company

Scientific Reasoning on Theorem QA

60.6Pass@1Critique-GRPO (Self-Critique)
20.0430.5741.151.63Jun 3, 2025

Evaluation results

MethodLinks
Critique-GRPO (Self-Critique)w/ External Supervision=true2025.0660.6
Critique-GRPO (CoT Critique)Backbone=Qwen3-8B2025.0660.1
Critique-GRPO (Self-Critique & Self-Evaluation)w/ External Supervision=false2025.0659.8
R1-GRPOw/ External Supervision=true2025.0657.9
SFTw/ External Supervision=true2025.0648.9
Qwen3-8B (w/ Think)2025.0646.9
Qwen3-8BBackbone=Qwen3-8B2025.0646.9
Critique-GRPO (CoT Critique)Backbone=Qwen2.5-7B-Base2025.0644
Qwen2.5-7B-BaseBackbone=Qwen2.5-7B-Base2025.0621.6