List your company

Reasoning on GPQA (Accuracy)

63.6AccuracyMOCHA
2.7618.55534.3550.145Sep 2, 2025Oct 15, 2025Nov 27, 2025Jan 9, 2026Feb 21, 2026Apr 5, 2026May 19, 2026

Evaluation results

MethodLinks
MOCHAEvaluation Backbone=Claude Haiku 4.5, Rollout Budget=1000, Number of Seeds=52026.0563.6
Seed SkillEvaluation Backbone=Claude Haiku 4.5, Rollout Budget=1000, Number of Seeds=52026.0559.2
TextGradEvaluation Backbone=Claude Haiku 4.5, Rollout Budget=1000, Number of Seeds=52026.0559.2
ProTeGiEvaluation Backbone=Claude Haiku 4.5, Rollout Budget=1000, Number of Seeds=52026.0559.2
GEPAEvaluation Backbone=Claude Haiku 4.5, Rollout Budget=1000, Number of Seeds=52026.0559.2
top-pModel=LLaMA3.1-8B-Instruct, Temperature=1.0, Few-shot=8-shot2025.0932.81
top-HModel=Phi-3-Mini, Temperature=1.0, Few-shot=8-shot2025.0932.37
min-pModel=Phi-3-Mini, Temperature=1.0, Few-shot=8-shot2025.0931.92
top-HModel=Phi-3-Mini, Temperature=1.5, Few-shot=8-shot2025.0930.8
top-HModel=Phi-3-Mini, Temperature=2.0, Few-shot=8-shot2025.0930.8
top-HModel=LLaMA3.1-8B-Instruct, Temperature=1.5, Few-shot=8-shot2025.0930.58
top-pModel=Phi-3-Mini, Temperature=1.0, Few-shot=8-shot2025.0930.58
min-pModel=Qwen2.5 3B, Temperature=1.5, Few-shot=8-shot2025.0930.13
min-pModel=Phi-3-Mini, Temperature=1.5, Few-shot=8-shot2025.0929.91
top-HModel=LLaMA3.1-8B-Instruct, Temperature=1.0, Few-shot=8-shot2025.0929.24
top-HModel=Qwen2.5 3B, Temperature=1.0, Few-shot=8-shot2025.0928.79
top-HModel=LLaMA3.1-8B-Instruct, Temperature=2.0, Few-shot=8-shot2025.0928.79
top-pModel=LLaMA3.1-8B-Instruct, Temperature=1.5, Few-shot=8-shot2025.0928.57
top-pModel=Phi-3-Mini, Temperature=1.5, Few-shot=8-shot2025.0928.57
min-pModel=Qwen2.5 3B, Temperature=1.0, Few-shot=8-shot2025.0928.35
min-pModel=LLaMA3.1-8B-Instruct, Temperature=1.5, Few-shot=8-shot2025.0928.35
top-HModel=Qwen2.5 3B, Temperature=2.0, Few-shot=8-shot2025.0928.12
top-HModel=Qwen2.5 3B, Temperature=1.5, Few-shot=8-shot2025.0927.9
top-pModel=Qwen2.5 3B, Temperature=1.0, Few-shot=8-shot2025.0927.68
top-pModel=Qwen2.5 3B, Temperature=1.5, Few-shot=8-shot2025.0927.23
min-pModel=LLaMA3.1-8B-Instruct, Temperature=1.0, Few-shot=8-shot2025.0926.34
min-pModel=LLaMA3.1-8B-Instruct, Temperature=2.0, Few-shot=8-shot2025.0926.12
min-pModel=Qwen2.5 3B, Temperature=2.0, Few-shot=8-shot2025.0925
top-pModel=LLaMA3.1-8B-Instruct, Temperature=2.0, Few-shot=8-shot2025.0923.88
min-pModel=Phi-3-Mini, Temperature=2.0, Few-shot=8-shot2025.0923.44
top-pModel=Qwen2.5 3B, Temperature=2.0, Few-shot=8-shot2025.0922.32
Llama3.2-3B-RLVRBackbone=Llama3.2-3B, Training Strategy=RLVR2026.0522.2
top-pModel=Phi-3-Mini, Temperature=2.0, Few-shot=8-shot2025.0918.53
Llama3.2-3B-GRLOBackbone=Llama3.2-3B, Training Strategy=GRLO2026.0517.7
Llama3.2-3B-GRLO+RLVRBackbone=Llama3.2-3B, Training Strategy=GRLO+RLVR2026.0517.7
Llama3.2-3B-InstructBackbone=Llama3.2-3B, Training Strategy=Instruct2026.0510.1
Llama3.2-3B-SFTBackbone=Llama3.2-3B, Training Strategy=SFT2026.055.1