paper-with-me

홈 › Papers

Agent-G^2: Gaussian Guidance for Agentic Reinforcement Learning

2026-08-24 · Zixuan Wang, Yanrui Miao, Zhengxi Lu, Teng Pan, Yiwen Qiu, Hongxing Li, Peng Qiu, Ruiqing Zhang, Yongliang Shen arxiv

Hint-based reinforcement learning addresses reward sparsity in long-horizon agentic tasks by retaining a prefix of an expert trajectory before each rollout, letting the policy explore from a state closer to success. Its effectiveness hinges on the guidance depth: how much of the trajectory to keep. Existing methods treat this depth as a deterministic scalar. Scheduled approaches share one value across samples and ignore per-task heterogeneity; per-sample probing estimates it separately at the cost of extra rollouts. We find that useful guidance occupies a band of depths whose informativeness profile is approximately Gaussian around the band center, rather than concentrating at a single optimal point. We propose Agent-G^2, a Gaussian guidance framework that draws the depth per task from a Gaussian whose center and spread are estimated online from rollouts already collected for policy optimization, requiring no probe rollouts or learned depth predictor. The center combines a global baseline with per-cluster difficulty, and the spread tracks within-cluster variance. We evaluate Agent-G^2 on ALFWorld and WebShop on Qwen2.5-1.5B / 7B-Instruct. Agent-G^2 outperforms the strongest hint-based, hint-free, and Aux-RL baselines on ALFWorld by 2.3 / 3.9 / 7.4 points at under one-third the rollout cost of per-sample probing.

📄 PDF Abstract BibTeX arXiv:2608.23318

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Learning Agentic Policy from Action Guidance

2026-05-12 · Yuxiang Ji, Zengbin Wang, Yong Wang, Shidong Yang 외 arxiv

Agentic reinforcement learning (RL) for Large Language Models (LLMs) critically depends on the exploration capability of the base policy, as training signals emerge only within its in-capability region. For tasks where t…

Reinforcement Learning

Group-Reflective Self-Distillation for Agentic Reinforcement Learning

2026-07-30 · Binbin Zheng, Zijun Xie, Guanqun Zhao, Enlei Gong 외 arxiv

Reinforcement learning with verifiable rewards (RLVR) is effective for training large language model agents. However, terminal rewards provide only coarse trajectory-level supervision, leaving successful behaviors, recur…

Reinforcement Learning

Exploring Reasoning Reward Model for Agents

2026-01-29 · Kaixuan Fan, Kaituo Feng, Manyuan Zhang, Tianshuo Peng 외 arxiv

Agentic Reinforcement Learning (Agentic RL) has achieved notable success in enabling agents to perform complex reasoning and tool use. However, most methods still relies on sparse outcome-based reward for training. Such …

Reinforcement Learning

Can Compact Language Models Search Like Agents? Distillation-Guided Policy Optimization for Preserving Agentic RAG Capabilities

2025-08-27 · Rikuto Kotoge, Mai Nishimura, Jiaxin Ma arxiv

Reinforcement Learning has emerged as a dominant post-training approach to elicit agentic RAG behaviors such as search and planning from language models. Despite its success with larger models, applying RL to compact mod…

Reinforcement Learning

Self-Distilled Agentic Reinforcement Learning

2026-05-14 · Zhengxi Lu, Zhiyuan Yao, Zhuowen Han, Zi-Han Wang 외 arxiv

Reinforcement learning (RL) has emerged as a central paradigm for post-training LLM agents, yet its trajectory-level reward signal provides only coarse supervision for long-horizon interaction. On-Policy Self-Distillatio…

Reinforcement Learning