List your company

Scientific Reasoning on GPQA Diamond (Pass@1, Token Count, Saved %)

2,004.7Token CountTRACES
1,803.93,159.34,514.75,870.1Apr 22, 2026

Evaluation results

MethodLinks
TRACESthreshold=0.9, Backbone=DS-Qwen14B, Seed=422026.042,004.755.871.46
TRACESthreshold=0.8, Backbone=DS-Qwen14B, Seed=422026.042,874.658.359.07
Basel. P^{(1)}_{system}Backbone=DS-Qwen14B, Seed=422026.043,736.253.946.81
TRACESthreshold=0.7, Backbone=DS-Qwen14B, Seed=422026.043,866.462.244.95
Basel. P^{(3)}_{system}Backbone=DS-Qwen14B, Seed=422026.044,530.156.435.51
TRACESthreshold=0.6, Backbone=DS-Qwen14B, Seed=422026.044,570.160.334.94
Basel. P^{(0)}_{system}Backbone=DS-Qwen14B, Seed=422026.044,768.357.132.12
Basel. P^{(0)}_{user}Backbone=DS-Qwen14B, Seed=422026.044,80157.731.66
TRACESthreshold=0.5, Backbone=DS-Qwen14B, Seed=422026.045,297.758.324.58
TRACESthreshold=0.4, Backbone=DS-Qwen14B, Seed=422026.045,683.258.919.1
StandardBackbone=DS-Qwen14B, Seed=422026.047,024.760.3—