paper-with-me

홈 › Papers

AGPO: Asymmetric Group Policy Optimization for Verifiable Reasoning and Search Ads Relevance at JD

2026-05-07 · Yang Xu, Kun Yao, Yiming Deng, Zheng Fang, Kai Ming Ting, Ming Pang arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) has demonstrated notable success in enhancing the reasoning performance of large language models (LLMs). However, recent studies reveal that while current RLVR methods improve sampling efficiency towards correct paths, they do not elicit fundamentally new reasoning patterns. Instead, the reasoning capability boundary of trained models often narrows compared to their base models, with base models achieving higher coverage at large sample sizes. In this work, we propose Asymmetric Group Policy Optimization (AGPO) to counteract this boundary shrinkage. AGPO adopts a negative-dominant reinforcement strategy to suppress incorrect reasoning paths, maintaining the base model's exploration capacity. For positive reinforcement, AGPO adopts a group advantage mechanism, which scales positive updates based on intra-group variance, allowing the model to focus on rare correct paths while suppressing updates from trivial paths. Our experiments on five mathematical benchmarks demonstrate that AGPO achieves state-of-the-art accuracy while consistently improving pass@$k$ performance at scale. In a large-scale industrial application for search ads relevance optimization, AGPO effectively enhances the quality of the data annotation, leading to substantial performance gains in downstream student models.

📄 PDF Abstract BibTeX arXiv:2605.05826

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

VAGPO: Vision-augmented Asymmetric Group Preference Optimization for Graph Routing Problems

2025-08-03 · Shiyan Liu, Bohan Tan, Zhiguang Cao, Yan Jin arxiv

Graph routing problems play a vital role in web-related networks, where finding optimal paths across graphs is essential for efficient data transmission and content delivery. Classic routing formulations such as the Trav…

AGPO: Adaptive Group Policy Optimization with Dual Statistical Feedback

2026-05-20 · Miaobo Hu, Shuhao Hu, Bokun Wang, Ruohan Wang 외 arxiv

Reinforcement learning improves LLM reasoning, but PPO/GRPO typically use fixed clipping and decoding temperature, which makes training brittle and tuning-heavy. We propose Adaptive Group Policy Optimization (AGPO), a cr…

Reinforcement Learning

GAGPO: Generalized Advantage Grouped Policy Optimization

2026-05-13 · Siyuan Zhu, Chao Yu, Rongxin Yang, Zongkai Liu 외 arxiv

Reinforcement learning has become a powerful paradigm for post-training large language model agents, yet credit assignment in multi-turn environments remains a challenge. Agents often receive sparse, trajectory-level rew…

Reinforcement Learning

Adaptive Group Policy Optimization: Towards Stable Training and Token-Efficient Reasoning

2025-03-20 · Chen Li, Nazhou Liu, Kai Yang

Since DeepSeek-R1 popularized, Group Relative Policy Optimization (GRPO) has become the core part of training Reasoning LLMs. However, we find some deficiency that influences RL stability and inference efficiency, like z…

Multi-Agent Guided Policy Optimization

2025-07-24 · Yueheng Li, Guangming Xie, Zongqing Lu arxiv

Due to practical constraints such as partial observability and limited communication, Centralized Training with Decentralized Execution (CTDE) has become the dominant paradigm in cooperative Multi-Agent Reinforcement Lea…

Multi-agent Reinforcement Learning