List your company

Mathematical Problem Solving on MATH 500 (Accuracy (avg@k))

97.83Accuracy (avg@k)TRACE (FKL on Ky)
81.08685.43389.7894.127May 11, 2026

Evaluation results

MethodLinks
TRACE (FKL on Ky)Backbone=Qwen3-8B, Symmetric Think evaluation=true2026.0597.83
TRACE (RKL on Ey)Backbone=Qwen3-8B, Symmetric Think evaluation=true2026.0597.48
GRPOBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0597.3
BaseBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0596.8
RLSDBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0596.68
SRPOBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0596.48
SDPOBackbone=Qwen3-8B, Symmetric Think evaluation=true2026.0595.13
TRACE (RKL on Ey)Backbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0592.05
BaseBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0591.35
RLSDBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0591.12
TRACE (FKL on Ky)Backbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0591
GRPOBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0590.33
SRPOBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0586.02
SDPOBackbone=Qwen3-1.7B, Symmetric Think evaluation=true2026.0581.73