List your company

STEM Question Answering on GPQA Main

0.181AccuracyQwen3-4B-Inst-2507
0.120680.136340.1520.16766Feb 5, 2026

Evaluation results

MethodLinks
Qwen3-4B-Inst-2507Chat Template=Off, Decoding=ConfAdapt, tau=0.62026.020.1812
L3.1-8B-MagpieChat Template=On, Decoding=Baseline, Step=0, k=12026.020.1761
L3.1-8B-MagpieChat Template=On, Decoding=ConfAdapt, tau=0.72026.020.1722.7
L3.1-8B-MagpieChat Template=On, Decoding=Static, k=12026.020.1541
Qwen3-4B-Inst-2507Chat Template=Off, Decoding=Baseline, Step=0, k=12026.020.1231