MDP-GRPO: Stabilized Group Relative Policy Optimization for Multi-Constraint Instruction Following

Mohammad Mahdi Salmani-Zarchi, Zahra Rahimi, Heshaam Faili, Mohammad Javad Dousti

Jun 5, 2026 at 04:00

11 Visningar

0 Kommentarer

arXiv:2606.06058v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards is ideal for multi-constraint instruction following, yet standard group-relative policy optimization (GRPO) becomes unstable under discrete, low-dispersion rewards, where within-group reward distributions are frequently homogeneous. We identify and...

Läs hela artikeln hos källan.

Läs originalartikeln

Var detta hjälpsamt?

Dela: