List your company

Mathematical Reasoning on Math-500 (Acc., Avg. Δ)

95.2AccuracyPaTaRM
-3.80821.89647.673.304Oct 28, 2025Nov 24, 2025Dec 22, 2025Jan 19, 2026Feb 16, 2026Mar 16, 2026Apr 13, 2026

Evaluation results

MethodLinks
PaTaRMPolicy Model=Qwen3-8B2025.1095.2—
Rule-based RewardPolicy Model=Qwen3-8B2025.1095—
Skywork-Llama-3.1-8BPolicy Model=Qwen3-8B2025.1093.6—
Base ModelPolicy Model=Qwen3-8B2025.1090—
PaTaRMPolicy Model=Qwen3-0.6B2025.1078—
Rule-based RewardPolicy Model=Qwen3-0.6B2025.1076.4—
Skywork-Llama-3.1-8BPolicy Model=Qwen3-0.6B2025.1074.2—
Base ModelPolicy Model=Qwen3-0.6B2025.1072.2—
DECOVECModel=Qwen2-7B, Protocol=Few-Shot (BM25)2026.0448.830.59
DECOVECModel=Qwen2-7B, Protocol=Few-Shot (KATE)2026.0443.365.49
DECOVECModel=Qwen2-7B, Protocol=Few-Shot (Rand. Std.)2026.0441.021.38
DECOVECModel=Gemma-2-9B, Protocol=Few-Shot (BM25)2026.0437.892.35
DECOVECModel=Llama-3-8B, Protocol=Few-Shot (KATE)2026.0437.112.96
DECOVECModel=Qwen2-7B, Protocol=Few-Shot (Rand. Ext.)2026.0436.332.15
DECOVECModel=Gemma-2-9B, Protocol=Few-Shot (KATE)2026.0435.942.36
DECOVECModel=Llama-3-8B, Protocol=Few-Shot (BM25)2026.0433.21.18
DECOVECModel=Gemma-2-9B, Protocol=Few-Shot (Rand. Ext.)2026.0426.561.38
DECOVECModel=Llama-3-8B, Protocol=Few-Shot (Rand. Ext.)2026.0421.882.16
Zero-ShotModel=Qwen2-7B, Protocol=Zero-Shot2026.0421.09—
DECOVECModel=Gemma-2-9B, Protocol=Few-Shot (Rand. Std.)2026.0421.095.49
DECOVECModel=Qwen2-1.5B, Protocol=Few-Shot (BM25)2026.0420.74.9
DECOVECModel=Qwen2-1.5B, Protocol=Few-Shot (KATE)2026.0419.925.5
DECOVECModel=Llama-3-8B, Protocol=Few-Shot (Rand. Std.)2026.0418.752.55
DECOVECModel=Qwen2-1.5B, Protocol=Few-Shot (Rand. Ext.)2026.0417.970.99
DECOVECModel=Qwen2-1.5B, Protocol=Few-Shot (Rand. Std.)2026.0417.582.56
DECOVECModel=Yi-6B, Protocol=Few-Shot (BM25)2026.0417.581.96
DECOVECModel=Yi-6B, Protocol=Few-Shot (KATE)2026.0417.193.54
DECOVECModel=Llama-2-7B, Protocol=Few-Shot (BM25)2026.0412.52.74
DECOVECModel=Qwen2-0.5B, Protocol=Few-Shot (BM25)2026.0410.550.79
DECOVECModel=Llama-2-7B, Protocol=Few-Shot (KATE)2026.049.773.91
DECOVECModel=Qwen2-0.5B, Protocol=Few-Shot (Rand. Std.)2026.048.592.37
DECOVECModel=Qwen2-0.5B, Protocol=Few-Shot (KATE)2026.048.594.72
DECOVECModel=Qwen2-0.5B, Protocol=Few-Shot (Rand. Ext.)2026.047.032.17
DECOVECModel=Yi-6B, Protocol=Few-Shot (Rand. Std.)2026.045.470.39
Zero-ShotModel=Gemma-2-9B, Protocol=Zero-Shot2026.045.47—
DECOVECModel=Yi-6B, Protocol=Few-Shot (Rand. Ext.)2026.045.082.45
Zero-ShotModel=Yi-6B, Protocol=Zero-Shot2026.043.52—
DECOVECModel=Llama-2-7B, Protocol=Few-Shot (Rand. Ext.)2026.042.731.57
DECOVECModel=Llama-2-7B, Protocol=Few-Shot (Rand. Std.)2026.041.951.57
Zero-ShotModel=Qwen2-0.5B, Protocol=Zero-Shot2026.041.17—
Zero-ShotModel=Qwen2-1.5B, Protocol=Zero-Shot2026.040.78—
Zero-ShotModel=Llama-2-7B, Protocol=Zero-Shot2026.040—
Zero-ShotModel=Llama-3-8B, Protocol=Zero-Shot2026.040—