HARTS is a training system that accelerates agentic reinforcement learning with hybrid‑attention models by sharing prefixes of irregular rollout trees. It uses a two‑level planner to create compression‑oriented microbatches, refines their membership and execution order, and accumulates gradients before a single backward pass through shared prefixes. The system supports distributed parallelism, activation recomputation, and preserves RL objectives while achieving 4.77‑10.45× higher update throughput on large hybrid‑attention MoE models.
By Boyuan Meng, Peihua Bao, Hong Liu, Xiaowei Zhu, Chao Wang, Gen Li, Zhenxuan Pan
The paper introduces OMAF, an Online MARL framework that uses a one-step Transformer-based flow policy to generate coordinated actions efficiently. By replacing costly iterative sampling with a single-step action generation and a joint optimization scheme that couples softmax Q-value estimation with flow policy objectives, OMAF maintains expressive multimodal behavior while improving training speed. Experiments on 10 tasks from MPE and MAMuJoCo demonstrate up to 3.4× higher returns and 10.5× better sample efficiency compared to baseline methods.
By Zhuoran Li, Yunzhan Li, Xun Wang, Yihan Du, Longbo Huang
arXiv:2602. 07339v2 Announce Type: replace Abstract: Diffusion-based trajectory planners can model multi-modal driving behavior, but their iterative denoising process introduces a latency bottleneck for real-time closed-loop deployment.
By Ruturaj Reddy, Hrishav Bakul Barua, Junn Yong Loo, Thanh Thi Nguyen, Ganesh Krishnasamy
arXiv:2510. 05592v2 Announce Type: replace Abstract: Outcome-driven reinforcement learning has advanced reasoning in large language models (LLMs), but prevailing tool-augmented approaches train a single, monolithic policy that interleaves thoughts and tool calls under full context; this scales poorly with long horizons and diverse tools and generalizes weakly to new scenarios.
By Zhuofeng Li, Haoxiang Zhang, Seungju Han, Sheng Liu, Jianwen Xie, Yu Zhang, Yejin Choi, James Zou, Pan Lu
arXiv:2611.17793v1 Announce Type: new
Abstract: The performance of offline Multi-Agent Reinforcement Learning (MARL) is limited by the quality of the fixed offline dataset, motivating trajectory augm...
By Jiwon Jeon, Myungsik Cho, Woojun Kim, Seongmin Kim, Woohyeon Byeon, Seungyul Han, Youngchul Sung
arXiv:2604. 10169v2 Announce Type: replace Abstract: Trajectory prediction is a key component of autonomous driving systems because future motions directly affect collision checking, behavior planning, and control.
By Wenchang Duan, Zhenguo Gao, Jinguo Xian, Yi Shi
arXiv:2603.02604v3 Announce Type: replace
Abstract: We introduce Heterogeneous Agent Collaborative Reinforcement Learning (HACRL), a new Reinforcement Learning from Verifiable Reward (RLVR) problem t...
By Zhixia Zhang, Zixuan Huang, Gongxun Li, Huaiyang Wang, Chengyi Yuan, Xin Xia, Deqing Wang, Fuzhen Zhuang, Shuai Ma, Ning Ding, Yaodong Yang, Jianxin Li, Yikun Ban
arXiv:2609.05837v1 Announce Type: new
Abstract: LLM-based agents are increasingly deployed in real-world applications through tool-use APIs, yet training them for specific environments remains fundam...
By Zhiyi Lyu, Yewen Li, Longtao Zheng, Shengtian Yang, Lang Feng, Lei Feng, Peng Jiang, Kun Gai, Qingpeng Cai, Bo An
CoSkill introduces a unified multi‑agent reinforcement learning framework that jointly trains a Reasoning Agent and a learnable Meta‑Skill Agent over a hierarchical skill library. By treating the meta‑skill workflow as a trainable agent and sharing a single backbone, CoSkill enables end‑to‑end co‑adaptation, allowing the Reasoning Agent to condition actions on retrieved task and step skills while the Meta‑Skill Agent refines those skills based on task performance. Experiments on ALFWorld and WebShop demonstrate that CoSkill outperforms prior skill‑based and RL baselines, achieving higher success rates and improved sample, asymptotic, and wall‑clock efficiency.
By Jinyuan Feng, Dongmin Li, Yiqun Chen, Yang Gao, Xing Chen, Huimu Wang, Zhiqiang Pu
arXiv:2604.05656v2 Announce Type: replace
Abstract: Vision-language-action (VLA) models with flow-matching action heads, such as {\pi}0 and {\pi}0.5, have become a standard recipe for generalist robo...
By Wuyang Luan, Junhui Li, Weiguang Zhao, Rui Ma, Tieru Wu
arXiv:2608. 02391v1 Announce Type: cross Abstract: Tool-using large language model (LLM) agents produce long, multi-turn trajectories, making gradient-based post-training memory-intensive.
By Zhiyuan Wang, Shengcai Liu, Jiahao Wu, Ning Lu, Hui Ouyang, Shaofeng Zhang, Haoze Lv, Ke Tang
arXiv:2607. 10836v1 Announce Type: new Abstract: Multi-agent ensembling multiplies active parameters and inference cost without answering three basic questions: which agents to consult, how deeply a query should traverse a hierarchy of agents, and when inter-agent communication is worth its cost.
By Sudipto Ghosh, Tanmoy Chakraborty