Denser neq Better: Limits of On-Policy Self-Distillation for Continual Post-Training
Paper • 2607.01763 • Published • 3
SDPO@CL
SDPO under Continual Learning
-
-
Moenupa/Qwen3-4B-Thinking-2507-SDPO0-MathChem
Updated -
Moenupa/Qwen3-4B-Thinking-2507-SDPO0-MathChemTool
Updated -
Moenupa/Qwen3-4B-Thinking-2507-SDPO0-MathChemToolCode
Updated -
Moenupa/Qwen3-4B-Thinking-2507-SDPO5-MathChem
Updated -
Moenupa/Qwen3-4B-Thinking-2507-SDPO5-MathChemTool
Updated -
Moenupa/Qwen3-4B-Thinking-2507-SDPO5-MathChemToolCode
Updated