Tag
SA-MRPO introduces a saturation-aware advantage reweighting technique for multi-reward policy optimization in reinforcement learning, improving performance on mathematical reasoning, adaptive reasoning, and coding benchmarks by focusing optimization on under-optimized objectives.
STARE addresses policy entropy collapse in GRPO-based reinforcement learning for large language models by introducing surprisal-guided token-level advantage reweighting and target-entropy regulation, achieving 4%-8% accuracy gains on AIME benchmarks.
GRAIL introduces gradient-reweighted advantages to improve token-level credit assignment in reinforcement learning for LLM reasoning, outperforming GRPO across multiple models.