MO-GRPO rescales each reward function by its group variance before summing them in GRPO, equalizing their influence and reducing reward hacking in multi-objective LLM training.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MO-GRPO: Mitigating Reward Hacking of Group Relative Policy Optimization on Multi-Objective Problems
MO-GRPO rescales each reward function by its group variance before summing them in GRPO, equalizing their influence and reducing reward hacking in multi-objective LLM training.