The paper introduces ReTRAIL, a framework that allows long‑horizon agents to rewind the environment to earlier states while preserving forward progress in knowledge. By training a shared policy with State‑Anchored GRPO and a branch‑advancement reward, ReTRAIL enables agents to recover from unproductive states and reuse insights from abandoned exploration. Experiments on MineSweeper, Sokoban, ALFWorld, and ScienceWorld show that this combined rewinding and experience‑reuse strategy outperforms simple rewind primitives and improves long‑horizon task solving.
By Ziang Ye, Wentao Shi, Lei Chen, Zhengzhou Cai, Yu Wang, Zhiyuan Yao, Zihan Wang, Xierui Song, Fuli Feng, Yixin Cao
A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, stale and off-path information consumes its limited capacity and obs...
arXiv:2610.20334v1 Announce Type: cross
Abstract: A ReAct-style agent accumulates observations, intermediate results, and failed attempts in a single context. As this context grows, stale and off-pat...
By Dongzhuo Li, Fangzhou Cheng, Weiwei Li, Kevin Han, Haixu Ma, Zhongchang Sun, Ruiyi Li, Serena Li, Xiangjun Fan, Lizhu Zhang, Yuandong Tian
arXiv:2607. 11388v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) and vision-language models (VLMs) have enabled increasingly capable digital agents for computer use.
By Wenyi Wu, Sibo Zhu, Kun Zhou, Aayush Salvi, Zixuan Song, Biwei Huang
The paper introduces OODA-Tool, a typed closed‑loop policy that separates state preservation from action generation to reduce state‑action competition in multi‑turn tool use. It follows Boyd’s Observe‑Orient‑Decide‑Act cycle, reconstructing task state, deciding on execution, forming admissible actions, and then realizing outputs. Experiments with Qwen3 models show OODA‑Tool consistently improves task success, especially for smaller models and tasks requiring accumulated information.
By Rongfeng Guo, Yinxuan Huang, Yusen Wu, Maoqing Zhong, Yunlu Chen, Meng Tang, Teng Long, Vincent Tao Hu
The paper introduces Trace, a framework that transforms sparse-reward trajectories into executable walkthroughs by identifying progress anchors, propagating credit, and estimating action prerequisites. Trace compiles noisy trajectories into state‑conditioned, verifiable procedures that remove loops and detours, enabling reuse, intermediate‑state resumption, and programmatic verification. Experiments on J‑TTL, WebShop, and ScienceWorld with three open‑source LLMs show that Trace outperforms eight baselines, improving average AUC and Final‑$3$ by 30.0% and 40.5% while using fewer inference tokens.
By Kaijie Chen, Chenyu Fang, Liang Yan, Bo Li, Bo Zhang, Peng Ye
arXiv:2610.14146v1 Announce Type: new
Abstract: Long-horizon terminal benchmarks measure whether an agent can sustain progress through extended workflows with process rewards, not only whether its fi...
By Zongxia Li, Zhongzhi Li, Yucheng Shi, Ruhan Wang, Junyao Yang, Chengsong Huang, Zhichao Liu, Xiyang Wu, Anhao Li, Yue Yu, Ergan Shang, Fuxiao Liu, Tong Zheng, Ninghao Liu, Lichao Sun, Haitao Mi, Leowei Liang
arXiv:2608. 04788v1 Announce Type: cross Abstract: Large language model agents are commonly trained through reinforcement learning with sparse trajectory-level rewards, which offer limited guidance on how strongly individual tokens should be updated.
By Yi Yang, Cong Qin, Xiaodan Liu, Chishui Chen, Qing Dong, Yan Zhang, Cao Liu, Zhao Yang, Lu Pan, Jiaye Lin, Yi Feng
The paper investigates how small action errors evolve when using action chunking in behavioural cloning. By injecting errors at each state and observing their growth under open‑loop (no replanning) and closed‑loop (replanning) regimes, the authors classify states as contracting, expanding, or unresolved. Across twelve manipulation tasks, they find that stable states are rare, error amplification is common, and that short‑horizon fitting can overestimate long‑horizon propagation. Predictors trained on camera and proprioceptive data can recover open‑loop stability but only partially capture closed‑loop dynamics, indicating that standard imitation learning does not reliably produce policies that contract errors when perturbed.
By Aryan Goyal
RePO‑VLA is a recovery‑driven policy optimization framework for Vision‑Language‑Action models that separates success, recovery, and failure trajectories. It introduces Recovery‑Aware Initialization, a Progress‑Aware Semantic Value Function, and Value‑Conditioned Refinement to learn from both successful and failed rollouts, enabling the policy to prefer high‑progress actions without online failure detectors. The authors also present the Failure‑Recovery Benchmark (FRBench) and demonstrate that RePO‑VLA raises adversarial success rates from 20% to 75% on average in simulated and real‑world bimanual tasks.
By Weijia Liufu, Xiaoyu Guo, Ruiyi Chen, Jingzhi Liu, Kaidong Zhang, Xiwen Liang, Jianqi Lin, Dawei Sun, Yuze Wang, Rongtao Xu, Bingqian Lin, Bowen Yang, Tongtong Cao, Bowen Peng, Dongyu Zhang, Guangrun Wang, Min Wang, Liang Lin, Xiaodan Liang
arXiv:2608. 06811v1 Announce Type: cross Abstract: Resolving a real software issue with a large language model (LLM) agent is a long repair episode, often tens to hundreds of steps spanning exploration, hypothesis, implementation, and verification.
By Jiahao Zhang, Yifan Zhang, Yu Huang
arXiv:2609.21423v1 Announce Type: new
Abstract: Online agent deployments produce abundant execution traces, while task-specific verification and expert annotation are costly to scale. We study how to...
By Siyuan Liu (Fudan University, Meituan Longcat Team), Fan Yu (Fudan University, Meituan Longcat Team), Dongyu Ru (Meituan Longcat Team), Yizhu Liu (Meituan Longcat Team), Yifan Yang (Meituan Longcat Team), Xuezhi Cao (Meituan Longcat Team), Xunliang Cai (Meituan Longcat Team), Yixin Cao (Fudan University)