method · first seen 4 Oct 2026
group relative policy optimization
Also: GMPO, GRPO, group-relative advantage estimation, reinforcement learning from group-relative policy optimization
Policy gradient method that estimates advantages by normalizing rewards across a group of sampled rollouts for the same prompt.
Group Relative Policy Optimization (GRPO) is a policy-gradient method that drops the learned value critic used in PPO. Instead, it samples a group of completions for each prompt and computes advantages by normalizing their rewards within that group. It was introduced by DeepSeek in 2024 and became a standard post-training recipe for reasoning models. The sources treat it as a default tool rather than a new idea. Papers apply it to video research agents, recommendation, red teaming, battery charging control and coding models. One paper ablates GRPO estimator choices at 1B scale. Repositories list it as a training stage alongside PPO, DAPO and SFT in agentic RL frameworks and distillation pipelines.
GRPO is already the common baseline for RL with verifiable rewards, so the spike in mentions reflects routine use across application papers rather than a new trend. The more useful signal to watch is the stream of variants and ablations that change its advantage estimator.
Analyst note (established), 17 Nov 2026.
Mentions per week
PapersCodeDiscussion
Rank 90 on 14 Nov 2026 · 179 mentions in 28 days · 172 independent groups · papers → code · first seen 6 weeks ago · acceleration -0.22
Earliest sources
- VideoResearchAgent: Grounded Task Synthesis and Sim-to-Real RL for Open-Web Video ResearchPapers · 4 Oct 2026 · Yuhang Zhou, Fei Li, Yuxi Wu, Bin Zhu, Jingjing Chen
- PWM: Personalized World Models with Online Reinforcement LearningPapers · 4 Oct 2026 · Zhexin Lou, Guancheng Lu, Zeyu Zhang, Yi Zhang, Yang Zhao, Hao Tang
- Prompt Dominance and Asymmetric Verifier Costs: Empirical Ablations of GRPO at 1B Scale on GSM8KPapers · 4 Oct 2026 · Yi Hou
- Learning to Simulate Individuals from Macro Social SignalsPapers · 5 Oct 2026 · Yining Zhao, Bushi Liu, Haofei Yu, Zhengyang Qi, Shanyong Wang, Chuyue Li, Yuxia…
- ThunderSyncRL: Lossless Acceleration of Agentic Reinforcement LearningPapers · 5 Oct 2026 · Seil Kang, Hangoo Kang, Tarun Suresh, Youngeun Kim, Shreyas Pimpalgaonkar, Seong…
- Transfer-Stratified On-Policy Distillation for RL-Improved Reasoning TeachersPapers · 5 Oct 2026 · Xiaoyu Chen, Bo Shao, Tiangang Zhu, Bintao Wu, Linjun Shou, Fengge Wu, Feng Sun,…
Latest sources
- ViRA: Toward Ultra-High-Resolution Visual Agents via Cross-Input Relative Policy OptimizationPapers · 16 Nov 2026 · Huawei Lin, Jinting Liu, Tongyan Wang, Xuyang Chen, Ruihua Shi, Zhichao Xu, Jiax…
- Learning to Reason with Predictive Gains for Generative RecommendationPapers · 16 Nov 2026 · Zhuoyang Li, Teng Shi, Jiakai Tang, Weijie Yu, Guofu Xie, Xiao Zhang, Lei Kou, Y…
- Stochastic Self-Distillation for Flow-Based Vision-Language-Action ModelsPapers · 16 Nov 2026 · Jingru Guo, Alessandra Russo, Joey Bose, Christos Ziakas
- AllocFlow: End-to-End Portfolio Trading Action Generation from Raw Market DataPapers · 15 Nov 2026 · Annan Wang, Yuchen Xie, Yu Cen, Leon Liangyu Chen, Qianyu Julie Zhu, Xin Li, Tia…
- Let Execution Do the TeachingPapers · 15 Nov 2026 · Abhishek Mishra, Sandeep Srinivasa, Arko Chattopadhyay, Enrique Martinho Ferrao,…
- Exploration of Reinforcement Learning in Generative Recommendation ModelsPapers · 15 Nov 2026 · Pengfei Zheng, Wuchao Li, Shiyao Wang, Honghui Bao, Hongyi Xie, Jiaxin Deng, Xin…
- Where Should LLM Agents Adapt at Test Time? In Context, in Weights, or BothPapers · 15 Nov 2026 · Parsa Mirtaheri, Mohammadreza Armandpour, Keivan Alizadeh, Iman Mirzadeh, Parshi…
- Attacker Diversity Reduction in Adaptive LLM Red TeamingPapers · 15 Nov 2026 · Chris Tava
- Large Language Model-Empowered Intelligent Agent for Adaptive Fast-Charging Control of Lithium-Ion BatteriesPapers · 13 Nov 2026 · Joey Chan, Haoyu Pan, Zhen Chen, ErShun Pan
- I2C-GRPO: Insight Importance Checklist for Reinforcement Learning in a Multi-Agent DDoS Alert Triage WorkflowPapers · 12 Nov 2026 · Eyal Zloof, Shaun Shuster, Aviram Zilberman, Chen Frydman, Inderjeet Singh, Andr…
- Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon AgentsPapers · 12 Nov 2026 · Haochen Liu, Qianhao Zhu, Zixuan Wang, Dingming Li, Tongbo Chen, Peng Qiu, Yingq…
- UAV-Flow++: A Unified Suite for Benchmarking and Learning Language-Guided UAV FlightPapers · 12 Nov 2026 · Donglin Yang, Xiangyu Wang, Yantong Zhong, Wenhao Zheng, Zihan Ding, Yue Liao, H…