List your company

Professional Reasoning on GPQA Diamond

52.5Accuracy (ACC)JET
29.41235.40641.447.394Sep 27, 2025

Evaluation results

MethodLinks
JETBackbone=DS-Qwen-7B2025.0952.55,530
DAPOBackbone=DS-Qwen-7B2025.09515,485
Laser-DBackbone=DS-Qwen-7B2025.09514,417
LCR1Backbone=DS-Qwen-7B2025.09503,429
DPOBackbone=DS-Qwen-7B2025.09495,519
BaseBackbone=DS-Qwen-7B2025.0947.56,359
Laser-DEBackbone=DS-Qwen-7B2025.0946.53,884
SFTBackbone=DS-Qwen-7B2025.0945.55,418
AdaptThinkBackbone=DS-Qwen-7B2025.0944.44,820
JETBackbone=DS-Qwen-1.5B2025.0943.44,182
DAPOBackbone=DS-Qwen-1.5B2025.0937.94,591
SFTBackbone=DS-Qwen-1.5B2025.0934.95,288
Laser-DBackbone=DS-Qwen-1.5B2025.0934.35,352
Laser-DEBackbone=DS-Qwen-1.5B2025.0933.85,575
LCR1Backbone=DS-Qwen-1.5B2025.0933.83,678
BaseBackbone=DS-Qwen-1.5B2025.0932.35,619
AdaptThinkBackbone=DS-Qwen-1.5B2025.0932.34,601
DPOBackbone=DS-Qwen-1.5B2025.0930.35,283