List your company

Mathematical Computation on GSM8K (EM, F1)

97.66EMPrompt-R1
29.404847.124964.84582.5651Nov 2, 2025

Evaluation results

MethodLinks
Prompt-R1Backbone=GPT-4o-mini2025.1197.6697.66
GRPOBackbone=Qwen3-4B2025.1192.9792.97
GEPAOptimization Framework=APO, Backbone=GPT-4o-mini2025.1187.590.1
BaselineBackbone=Qwen3-4B2025.1184.3884.38
CoT ReasoningBackbone=GPT-4o-mini2025.1184.3888.02
BaselineBackbone=GPT-4o-mini2025.1183.5986.72
CoT ReasoningBackbone=Qwen3-4B2025.1182.8182.81
TextGradOptimization Framework=APO, Backbone=GPT-4o-mini2025.1170.3185.99
OPROOptimization Framework=APO, Backbone=GPT-4o-mini2025.1163.2883.65
SFTBackbone=Qwen3-4B2025.1132.0332.03