List your company

Science Question Answering on GPQA Diamond (avg@k)

58.33Accuracy (avg@k)TRACE (FKL on Ky)
26.173234.521642.8751.2184May 11, 2026

Evaluation results

MethodLinks
TRACE (FKL on Ky)Backbone=Qwen3-8B, Symmetric Think evaluation=true2026.0558.33
BaseBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0558.27
RLSDBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0557.26
TRACE (RKL on Ey)Backbone=Qwen3-8B, Symmetric Think evaluation=true2026.0556.44
GRPOBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0553.85
SDPOBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0539.9
RLSDBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0538.89
TRACE (RKL on Ey)Backbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0538.67
SRPOBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0537.31
BaseBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0537.18
GRPOBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0537.18
TRACE (FKL on Ky)Backbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0536.77
SRPOBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0535.82
SDPOBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0527.41