List your company

Mathematical Reasoning on DAPO-Math In-Distribution (test)

84.1Avg@16FwdKL
48.2257.53566.8576.165Jul 7, 2026

Evaluation results

MethodLinks
FwdKLBackbone=Qwen3-8B2026.0784.1
GEOSDBackbone=Qwen3-8B2026.0782.7
SkewKLBackbone=Qwen3-8B2026.0782
RevKLBackbone=Qwen3-8B2026.0781.8
TrOPDBackbone=Qwen3-8B2026.0781.6
GRPOBackbone=Qwen3-8B2026.0780.9
SFTBackbone=Qwen3-8B2026.0780.8
JSDBackbone=Qwen3-8B2026.0780.7
RevKLBackbone=Olmo-3-7B-Think2026.0777.2
TIPBackbone=Qwen3-8B2026.0776.9
GEOSDBackbone=Olmo-3-7B-Think2026.0775.5
FwdKLBackbone=Olmo-3-7B-Think2026.0775.1
TrOPDBackbone=Olmo-3-7B-Think2026.0774.9
GRPOBackbone=Olmo-3-7B-Think2026.0774.7
SFTBackbone=Olmo-3-7B-Think2026.0773.8
TIPBackbone=Olmo-3-7B-Think2026.0773.6
JSDBackbone=Olmo-3-7B-Think2026.0773.1
SkewKLBackbone=Olmo-3-7B-Think2026.0770.6
FwdKLBackbone=DS-R1-Llama-8B2026.0770.1
GEOSDBackbone=DS-R1-Llama-8B2026.0769.4
RevKLBackbone=DS-R1-Llama-8B2026.0768.3
JSDBackbone=DS-R1-Llama-8B2026.0766.7
TIPBackbone=DS-R1-Llama-8B2026.0766
TrOPDBackbone=DS-R1-Llama-8B2026.0765.2
SkewKLBackbone=DS-R1-Llama-8B2026.0765.1
GRPOBackbone=DS-R1-Llama-8B2026.0762.5
SFTBackbone=DS-R1-Llama-8B2026.0761.8
π0Backbone=Qwen3-8B2026.0761.4
π0Backbone=Olmo-3-7B-Think2026.0753.3
π0Backbone=DS-R1-Llama-8B2026.0749.6