DeliveryGym is a 3D reinforcement learning environment that simulates continuous courier shifts, integrating multimodal tool interaction and persistent world dynamics to compute trajectory rewards based on simulator events. It provides feedback on resource consumption—time, energy, and money—across entire delivery trajectories, enabling agents to learn planning that balances immediate task success with long‑term constraints. The environment also adapts future training shifts to a policy’s weaknesses while keeping evaluation fixed, demonstrating that both learning from complete shifts and targeted training improve agent performance on complex delivery tasks.
arXiv:2509. 21842v2 Announce Type: replace Abstract: Travel planning (TP) agent has recently worked as an emerging building block to interact with external tools/resources for travel itinerary generation, ensuring an enjoyable user experience.
By Yansong Ning, Rui Liu, Jun Wang, Kai Chen, Wei Li, Jun Fang, Kan Zheng, Naiqiang Tan, Hao Liu
The paper introduces VHD-Play, a pipeline that first samples and solves a mathematical model before generating agentic reinforcement learning environments, ensuring that dynamics and evaluation are aligned from the outset. This approach yields 3,300 diverse environments at a low cost and significantly improves the performance of a large language‑model agent (Qwen3.6‑35B‑A3B) across multiple diagnostic families and external benchmarks. The study demonstrates that stateful interaction is a key factor in learning gains and that scaling the training substrate can further enhance performance.
By Xinjie Shen, Wei Fan, Xudong Guo, Jianhong Tu, Yang Su, Chuqiao Kuang, Yinger Zhang, Dayiheng Liu
arXiv:2606. 19990v1 Announce Type: new Abstract: While RL has become a promising tool for refining world models, existing methods largely rely on conservative rollouts near the training distribution, limiting exploration, behavioral diversity, and richer dynamic discovery.
By Pu Li, Zhigang Lin, Qiang Wu, Yongxuan Lv, Fei Wang, Shan You
The paper introduces Instruct-to-Act, a system that decouples planning and control by combining a vision‑language model (VLM) planner with a world‑model controller. The VLM generates sparse, high‑level text instructions, while the controller executes them at high frequency, trained via relabeling rollouts with synthetic instructions and joint optimization of behavior cloning, reward, and world‑model objectives. Across seven embodied environments—including multi‑agent settings—this approach outperforms controller‑only and direct VLM action methods, maintains fast control, and allows swapping pretrained VLM planners without fine‑tuning, achieving competitive results with strong baselines on most tasks.
By Zineng Tang, Kelsey R. Allen, Sjoerd van Steenkiste, Ishita Dasgupta, Alane Suhr
The paper introduces Imagine-then-Plan (ITP), a framework that lets agents learn by interacting with a learned world model to generate multi-step imagined trajectories. ITP features an adaptive lookahead mechanism that balances ultimate goals with task progress, producing richer signals about future outcomes. Experiments on various benchmarks show that ITP outperforms existing baselines, and analyses suggest the adaptive lookahead improves reasoning for complex tasks.
By Youwei Liu, Jian Wang, Hanlin Wang, Beichen Guo, Wenjie Li
arXiv:2608. 11338v1 Announce Type: cross Abstract: Recently, the practice of augmenting LLM agent capability with skills has gained prevalence.
By Zixi Huang, Xiheng Wang, Andrew Wang, William Jurayj, Bernal Jim\'enez Guti\'errez, Daniel Khashabi, Nicholas Andrews
The paper introduces Instruct-to-Act, a system that decouples high‑level planning from low‑latency control by combining a vision‑language model (VLM) planner with a world‑model controller. The VLM generates sparse, high‑level text instructions, while the controller executes them autonomously at high frequency. Experiments across seven embodied environments, including multi‑agent settings, show that this approach outperforms both controller‑only and direct VLM action‑generation methods, maintains fast control, and allows swapping in different pretrained VLM planners without fine‑tuning.
Teach-and-Grow Learning (TGL) is an agent-centered architecture that transforms a few successful demonstrations into reusable Skill Blocks, enabling a robot to compose, execute, and revise behaviors in new scenes without task-specific policy retraining. The system maintains a Skill Library and structured Experience Memory to capture successes, failures, and repairs, allowing persistent reuse and agent-directed adaptation. Evaluation on the LIBERO benchmark shows state-of-the-art performance, and the authors propose a scaling-law hypothesis suggesting that accumulated reusable experience reduces future-task error and teaching demand following a power-law trend.
By Chang Nie, Zhe Liu, Hesheng Wang
Continually improving agents require dynamic interaction feedback beyond static supervision, yet direct real-environment interaction is costly, slow, unsafe, and hard to parallelize. World modeling offers a natural intermediate proxy that allows agents to query lower-cost, more controllable feedback before committing to real actions.
arXiv:2608. 08523v1 Announce Type: new Abstract: Multimodal embodied agents are increasingly required to solve long-horizon tasks by integrating visual observations, textual goals, and interaction history into closed-loop decision making.
By Pengfei Xu, Yong Liu, Xiaoya Nan, Qiang Yang, Peilan Xu
Reinforced Planning with Latent World Models introduces RP1, a neural planner that learns to evaluate imagined outcomes via a critic and improve multi‑step plans through an optimizer trained offline on world‑model roll‑outs. Unlike existing planners that are hand‑designed or only inform policies, RP1 fully learns to refine plans and can be attached to any pretrained latent world model. In experiments on visual navigation, arm reaching, and robotic manipulation, RP1 outperforms hand‑designed search algorithms, achieving near‑perfect success while using 1,000× fewer roll‑outs and up to 67× faster inference.