List your company

Mathematical Reasoning on GSM8K zero-shot (test)

89.46AccuracyEPICAR
0.051223.263146.47569.6869Dec 12, 2025Dec 17, 2025Dec 22, 2025Dec 27, 2025Jan 1, 2026Jan 6, 2026Jan 11, 2026

Evaluation results

MethodLinks
EPICARFamily=Qwen-3-8B, Zero-shot=true2026.0189.460.7220.3640.216
EPICARFamily=Qwen-3-4B, Zero-shot=true2026.0188.250.7560.1320.106
STaRFamily=Qwen-3-8B, Zero-shot=true2026.0187.950.6780.2150.147
STaRFamily=Qwen-3-4B, Zero-shot=true2026.0186.050.6930.0990.122
Base ModelFamily=Qwen-3-8B, Zero-shot=true2026.0185.90.6540.2150.162
Base ModelFamily=Qwen-3-4B, Zero-shot=true2026.0185.060.6110.3640.255
EPICARFamily=Qwen-3-1.7B, Zero-shot=true2026.0179.30.660.6710.61
STaRFamily=Qwen-3-1.7B, Zero-shot=true2026.0178.540.4930.5410.463
Base ModelFamily=Qwen-3-1.7B, Zero-shot=true2026.0177.330.490.4450.376
VOYAGER|D_train|=1000, Evaluation Model=Gemma 7B-IT, Inference Mode=Zero-shot2025.1245.7———
VOYAGER|D_train|=500, Evaluation Model=Gemma 7B-IT, Inference Mode=Zero-shot2025.1242.8———
EPICARFamily=Llama-3-8B, Zero-shot=true2026.0137.450.5650.2230.284
DEFAULT|D_train|=1000, Evaluation Model=Gemma 7B-IT, Inference Mode=Zero-shot2025.1235.7———
STaRFamily=Llama-3-8B, Zero-shot=true2026.0132.150.5610.3380.332
Base ModelFamily=Llama-3-8B, Zero-shot=true2026.0131.390.5650.3290.324
EPICARFamily=Llama-3-3B, Zero-shot=true2026.0121.460.6060.020.166
STaRFamily=Llama-3-3B, Zero-shot=true2026.0117.130.4970.2840.228
VOYAGER|D_train|=1000, Evaluation Model=Gemma 2B-IT, Inference Mode=Zero-shot2025.1216.4———
Base ModelFamily=Llama-3-3B, Zero-shot=true2026.0114.940.5270.3030.223
DEFAULT|D_train|=1000, Evaluation Model=Gemma 2B-IT, Inference Mode=Zero-shot2025.1213.1———
VOYAGER|D_train|=500, Evaluation Model=Gemma 2B-IT, Inference Mode=Zero-shot2025.1212.9———
EPICARFamily=Llama-3-1B, Zero-shot=true2026.013.640.6450.9220.887
Base ModelFamily=Llama-3-1B, Zero-shot=true2026.013.490.5450.8410.741
STaRFamily=Llama-3-1B, Zero-shot=true2026.013.490.4780.840.74