Reinforcement learning for reasoning

PPO and RLOO for language modelsZiegler et al., 2019; Huang et al., 2024; Ahmadian et al., 20248 questions GRPO and DeepSeek-R1Shao et al., 2024; Guo et al., 20258 questions DAPO and Dr. GRPOYu et al., 2025; Liu et al., 20258 questions Group Sequence Policy OptimizationZheng et al., 20257 questions Tricks or traps in RL for reasoningLiu et al., 20258 questions Tülu 3 post-training recipeLambert et al., 20248 questions Let's Verify Step by StepLightman et al., 20248 questions Entropy collapse in RL for reasoningCui et al., 2025; Wang et al., 20258 questions RL and the set of solvable problemsYue et al., 2025; Liu et al., 20258 questions Compute-optimal test-time scalingSnell et al., 20258 questions Reinforcement learning for codeGehring et al., 2025; Wei et al., 20258 questions LLMs are greedy agentsSchmied et al., 20268 questions Single Deep CFR and DREAMSteinberger, 2019; Steinberger et al., 20208 questions Stochastic softmax tricks and Gumbel-RaoPaulus et al., 2020; Paulus et al., 20218 questions