The paper introduces Guidance-Augmented GRPO (GA‑GRPO), a theoretical framework that unifies several external‑guidance methods for reinforcement learning with verifiable rewards (RLVR) used to improve large language model reasoning. GA‑GRPO treats external guidance as a stochastic operator that rewrites the question distribution, yielding a biased on‑policy policy‑gradient estimator whose bias is bounded by the total‑variation guidance divergence. The authors prove convergence rates, derive an optimal guidance‑weight formula, and validate their predictions experimentally on a math‑reasoning benchmark, showing that the optimal‑weight GA‑GRPO outperforms existing methods while saving GPU time.
arXiv:2606. 16733v1 Announce Type: new Abstract: Policy gradient algorithms for language models optimize the same objective $J(\theta) = \mathbb{E}*{\tau \sim p*\theta(\tau)}[R(\tau)]$, which has exactly two factors: the trajectory probability $p_\theta(\tau)$ and the reward $R(\tau)$.
arXiv:2607. 27610v1 Announce Type: new Abstract: Reinforcement learning (RL) finetuning significantly enhances the reasoning capabilities of large language models (LLMs), yet its effectiveness critically depends on selecting prompts of appropriate difficulty for the current policy.
arXiv:2606. 11709v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense, token-level supervision for reasoning models by aligning a model's own distribution with the distribution it produces under privileged context, typically a verified solution.
The paper investigates why Group Relative Policy Optimization (GRPO) benefits from per‑prompt normalization by examining the local curvature of the sequence‑level policy gradient. It shows that standard deviation normalization acts as an adaptive gradient, yielding a provably faster convergence rate than unnormalized REINFORCE under mild conditions, with the improvement tied to the average within‑prompt reward standard deviation. The authors also propose IS‑GRPO, an importance‑sampling variant that maintains alignment with the full gradient and offers a tighter convergence guarantee, and empirically validate these theoretical insights on GSM8K and MATH datasets at 1.5B and 7B model scales.
arXiv:2601. 03895v2 Announce Type: replace-cross Abstract: Group Relative Policy Optimization (GRPO) has emerged as a popular algorithm for reinforcement learning with large language models (LLMs).
arXiv:2504. 18587v2 Announce Type: replace-cross Abstract: Reinforcement learning has emerged as a powerful approach for improving the reasoning capabilities of large language models, as demonstrated by systems such as OpenAI's O1~\cite{o1} and DeepSeek-R1~\cite{r1}.
arXiv:2606. 05434v1 Announce Type: new Abstract: Group Relative Policy Optimisation (GRPO) has emerged as an effective reinforcement-learning algorithm for aligning language models on reasoning tasks, but it treats every token position and every sampled rollout symmetrically.
arXiv:2609.26355v1 Announce Type: new Abstract: Reinforcement learning has become a central component of large language model (LLM) post-training, yet token-level credit lacks a generally accepted ma...
arXiv:2606. 18810v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has driven substantial progress in training LLMs for reasoning tasks, but representative methods such as GRPO assign uniform credit across all tokens, wasting gradient on routine tokens while under-crediting pivotal reasoning steps.
arXiv:2609.38018v1 Announce Type: new Abstract: Group relative policy optimization (GRPO) learns only from prompts whose sampled responses disagree: a group that is entirely correct or entirely incor...
arXiv:2606. 30789v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) has become a standard tool for improving the reasoning ability of large language models, yet its training dynamics are still described empirically: reward trajectories are fit with low-parameter functional forms whose constants carry no mechanistic meaning, and hyperparameter choices remain a matter of trial and error.
The paper introduces SORL, a framework that stabilizes off‑policy reinforcement learning for long‑horizon large language model agents. It identifies two key instability sources—token‑level policy granularity mismatches and high‑variance off‑policy updates—and proposes turn‑level importance sampling and clipping‑triggered normalization to align optimization with multi‑turn interactions. Two instantiations, SO‑PPO and SO‑GRPO, are evaluated on open‑domain, multi‑hop, and medical QA benchmarks, as well as on asynchronous RL for mathematical reasoning, showing improved robustness without the need for early stopping or heuristic tuning.