The paper introduces SPACE, a method for enabling large language model agents to emit variable-length action chunks in long-horizon tasks. By distilling chunk-boundary supervision from programmatic skills derived from successful trajectories, SPACE overcomes the tendency of agents to either act one step at a time or commit to overly long sequences. Experiments on ALFWorld and ScienceWorld demonstrate that SPACE raises success rates by 7.0%–31.3% and cuts LLM decision rounds by up to 78.9%.
By Yanting Yang, Can Jin, Jinman Zhao, Jiahao Wu, Yang Zhou, Zhepeng Wang, Zhendong Wang, Mu Zhou, Dimitris N. Metaxas
FlexiWorld is a JEPA-based latent world model that learns variable‑length action chunks across multiple time scales for goal‑directed planning. It jointly trains a causal action encoder and an autoregressive actor, using mixed‑span goal supervision and Student Forcing to reduce exposure bias. In experiments on four benchmarks, FlexiWorld with the Actor‑Residual Cross‑Entropy Method (ARCEM) achieves higher mean success rates than the strongest baseline and supports flexible planning chunk lengths without retraining.
By Shidu Ren, Qilin Gu, Zhenghao Ni, Junhan Sun, Jiaqi Wang, Damien Scieur, Yunze Liu
arXiv:2606. 10507v1 Announce Type: new Abstract: While Large Language Models (LLMs) have demonstrated strong capabilities as autonomous agents across a wide range of tasks, their performance often degrades in multi-turn long-horizon agentic tasks.
By Juncheng Diao, Zhicong Lu, Peiguang Li, Yongwei Zhou, Changyuan Tian, Qingbin Li, Rongxiang Weng, Jingang Wang, Xunliang Cai
arXiv:2609.36250v1 Announce Type: new
Abstract: Action chunking provides temporal abstraction in reinforcement learning by selecting short action sequences instead of individual actions, but many exi...
By Sanghyun Hahn, Jonghyun Choi
arXiv:2609.36471v1 Announce Type: cross
Abstract: World-Action Models (WAMs) improve robotic manipulation by conditioning action generation on predicted future observations, but future prediction add...
By Guoheng Sun, Chen Chen, Jin Wang, Ang Li, Teresa Lv
CoDA is a framework that lets a simulation data engine and an action policy evolve together using a shared memory of environment construction, policy failures, and learning progress. The memory stores validated diagnoses from a vision‑language critic, which are turned into generation programs that recreate challenging scenarios and training programs that help the policy improve. Using this approach, CoDA boosts out‑of‑distribution success on several manipulation tasks, raising success rates from 5.3 % to 27.6 % and achieving up to 84 % success on LIBERO‑Pro under perturbations.
By Tianjiao Yu, Zhilin Ou, Yifan Shen, Xinzhuo Li, Yuanzhe Liu, Onkar Kishor Susladkar, Fan Wang, Ismini Lourentzou
The paper introduces Predictive Action Chunk Learning (PACL), a method for improving robot manipulation policies using mixed-quality deployment experience. PACL first trains a predictive chunk-level critic to evaluate temporally extended action sequences, then uses the critic’s quality estimates to guide a diffusion actor that learns from both successful and failed rollouts. Experiments on simulated and real robots demonstrate that PACL consistently enhances pretrained policies and outperforms strong imitation learning and offline reinforcement learning baselines.
By Yangang Ren, Yujie Yan, Zirui Li, Jiaming Guo, Di Zeng, Ji Tao, Lan Yu, Xuesong Tian, Chen Lv
Natural Language Actor-Critic (NLAC) is a new actor‑critic algorithm that trains large language model (LLM) agents by using a generative LLM critic to produce natural‑language feedback instead of scalar rewards. By providing richer, actionable explanations for suboptimal actions, NLAC helps LLM policies improve without relying on random exploration, and it can be trained off‑policy without policy gradients. Experiments on reasoning, web browsing, tool‑use, and dialogue tasks show that NLAC outperforms existing training methods and offers a more scalable and stable training paradigm for LLM agents.
By Joey Hong, Kang Liu, Zhan Ling, Jiecao Chen, Sergey Levine
The paper introduces Rollout-Time Intervention (RTI), a method that uses reward-side evidence during rollouts to transform unsuccessful trajectories into successful ones, thereby providing learning signals for tasks beyond a policy’s intrinsic exploration limits. RTI filters out trajectories with excessive distributional shifts, allowing these off-policy interventions to guide policy optimization effectively. Experiments on five benchmarks show that RTI consistently outperforms existing RL baselines, enabling post-trained policies to approach the performance of stronger large language models and surpass their original capability boundaries.
By Chen Xi, Ding Ruixue, Wang Zhigaoyuan, Zhang Fanrui, Zhang Qiang, Chen Boli, Wang Shihang, Wang Tianfu, Ren Ailin, Wang Pengyong, Xie Pengjun, Xiong Hui
Agentic robotics places a large language model (LLM) agent in the robot control loop, where it composes perception, motion, and vision-language-action (VLA) tools under the guidance of a memory and th...
arXiv:2603. 01891v2 Announce Type: replace Abstract: Action chunking improves exploration and accelerates value propagation in long-horizon reinforcement learning, but naively applying off-policy methods to the temporally extended action space at reduced decision frequency offsets these gains, leading to poor sample efficiency.
By C. F. Maximilian Nagy, Onur Celik, Emiliyan Gospodinov, Florian Seligmann, Weiran Liao, Aryan Kaushik, Gerhard Neumann
arXiv:2610.18874v1 Announce Type: cross
Abstract: Agentic robotics places a large language model (LLM) agent in the robot control loop, where it composes perception, motion, and vision-language-actio...
By Bingxin Xu, Wenrui Bao, Xinxin Liu, Jiawei Zhou, Yuzhang Shang, Emilio Ferrara