DCPO combines dynamic clipping bounds and cumulative advantage standardization, achieving higher average math benchmark scores and higher response utilization than GRPO, DAPO, and GSPO.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DCPO: Dynamic Clipping Policy Optimization
DCPO combines dynamic clipping bounds and cumulative advantage standardization, achieving higher average math benchmark scores and higher response utilization than GRPO, DAPO, and GSPO.