List your company

Card Prediction on HigherLower 100 episodes (test)

0.53RewardSLM
0.49880.50690.5150.5231Jul 2, 2026

Evaluation results

MethodLinks
SLMModel=4B2026.070.5374100—
ASK+Model=2B2026.070.5274696
ASK+Model=4B2026.070.5274605
SLMModel=2B2026.070.5173100—
PPOModel=—2026.070.5720—
ASKModel=2B2026.070.573690
ASKModel=4B2026.070.572600