arXiv:2602.05863v3 Announce Type: replace
Abstract: Group Relative Policy Optimization (GRPO) remains the dominant critic-free approach for fine-tuning LLMs and VLMs, but its compatibility with const...
By Roger Girgis, Rodrigue de Schaetzen, Luke Rowe, Azal\'ee Robitaille, Christopher Pal, Liam Paull
arXiv:2509. 22047v3 Announce Type: replace Abstract: Group Relative Policy Optimization (GRPO) has been shown to be an effective algorithm when an accurate reward model is available.
By Yuki Ichihara, Yuu Jinnai, Tetsuro Morimura, Mitsuki Sakamoto, Ryota Mitsuhashi, Eiji Uchibe
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.
By Cheng Ge, Caitlyn Heqi Yin, Hao Liang, Jiawei Zhang
arXiv:2609.21432v1 Announce Type: new
Abstract: Post-training plays a pivotal role in enhancing the reasoning capabilities and task-specific expertise of large language models (LLMs). Despite recent...
By Kaichen Zhang, Yuzhong Hong, Junwei Bao, Hongfei Jiang, Yang Song, Dingqian Hong, Hui Xiong
arXiv:2606. 05800v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) often adopts GRPO-style group-relative updates, sampling multiple rollouts per prompt to construct normalized learning signals.
By Powei Chang, Jinpeng Zhang, Chaoqun Sun, MiniWell Tsao, Lianrui Li, Jianxiang Xiang, Chenyu Wang, Yukang Gao, Dongying Kong
arXiv:2609.39757v1 Announce Type: cross
Abstract: Black-box distillation is a practical route for transferring capabilities from API-accessible large language models that expose only text outputs int...
By Xiao Cui, Mo Zhu, Yulei Qin, Yuze Wu, Wengang Zhou, Houqiang Li