HorizonFlow is a hierarchical planner for offline goal-conditioned reinforcement learning that treats the planning horizon as an output rather than a fixed input. It uses a subgoal route planner and an action-prefix controller, both employing insertion-based generation and flow matching, to jointly generate continuous plan content and its length. The method leverages the partially generated plan to guide token insertion and to steer generation toward shorter plans, achieving superior performance on Maze2D, Multi2D, and OGBench benchmarks.
By JunHyeok Oh, Zian Jang, Byung-Jun Lee
The paper introduces Hamiltonian Temporal Contrastive Curiosity (HTC²), a method that combines multi‑horizon temporal representation learning with Hamiltonian phase‑space structure to guide exploration in physical control environments. HTC² uses a horizon‑conditioned contrastive critic to learn relationships between future state pairs within an episode, and the resulting representation distance serves as an intrinsic reward. The reward is further modulated by bounded multi‑step Hamiltonian variation, and the framework is integrated with Soft Actor‑Critic and Proximal Policy Optimization, evaluated on six DeepMind Control Suite tasks against several baseline curiosity methods.
By Shuwei Wang, Qiang Guan, Brian Qiu, Haoxiang Huang, Haoxi Wang, Chenyu Li, Bo Xu, Sen Cui
Exploration in physical control environments requires agents to identify actions whose consequences reveal previously unexplored aspects of the underlying dynamics. However, many intrinsic motivation...
Recent advances in generative planning have made trajectory inpainting a promising approach to offline goal-conditioned reinforcement learning. However, these methods typically specify the planning ho...
The paper introduces Planning Under Noise (PUN), a neural operator that learns noise‑aware value fields for planning in unseen geometries. By imposing reflecting Neumann conditions on obstacles and a Dirichlet condition at the goal, PUN allows sampled paths to explore free space until reaching the target. It extends the Monte Carlo solver Walk on Stars to higher dimensions, learns geometry‑to‑field mappings from noisy labels, and achieves high success rates and low runtimes on 4‑D replanning benchmarks across multiple noise levels.
By Michael Chen, Hong Chul Nam, Xi Deng, Anima Anandkumar
The paper introduces the Dual-Latent World Model (Dual-WM), which separates local execution and long-range planning into distinct latent spaces and dynamics models. A new learning method, Long-Horizon Representation Learning with Weighted Rollout (LoRe), supervises predictions at both levels using exponential horizon weights. Experiments on five goal-conditioned visual control tasks show that Dual-WM improves success rates over strong baselines, especially at longer horizons.
By Delin Zhao, Zhengrong Yue, Shaobin Zhuang, Junlin He, Xiaoyu Chen, Zikang Wang, Yuxin Liu, Limin Wang, Yali Wang
arXiv:2610.17927v1 Announce Type: cross
Abstract: Autonomous navigation in real-world environments often requires long-horizon operation under partial observability and evolving goals. Existing maple...
By Yousef Liang, Ryan Ardalan, Yarden Goraly, Chenhao Zhu, Yue Yin, Jonas Frey, Claire Tomlin, Marco Pavone, Shankar Sastry
The paper presents a deep active inference framework for real‑world robotic navigation that combines a diffusion policy with a multiple‑timescale recurrent state‑space model. The diffusion policy generates diverse candidate actions, while the state‑space model predicts long‑horizon outcomes, allowing the system to select actions that minimize expected free energy. Experiments show higher success rates and fewer collisions, especially in exploration‑heavy scenarios, demonstrating the effectiveness of this unified exploration and goal‑directed approach.
By Riko Yokozawa, Kentaro Fujii, Yuta Nomura, Shingo Murata
The paper introduces LG-DKGP, a locality‑gated deep‑kernel Gaussian‑process prior that learns action continuations from the temporal structure of expert demonstrations. By combining recent and longer‑range action history with a neural mean and a GP residual, the planner corrects action proposals and initializes exploration around a learned mean. Evaluated on five simulated tasks, the method attains 87.07% macro success, outperforming prior planners such as INTACT, PRISM, and LeWM.
arXiv:2609.34575v2 Announce Type: replace
Abstract: Offline goal-conditioned reinforcement learning (GCRL) learns goal-directed policies from reward-free data, but in long-horizon tasks, goal-conditi...
By Hengrui Zhang, Yuhu Cheng, C. L. Philip Chen, Xuesong Wang
The paper introduces Hindsight Inverse JEPA Planning (HIJP), a search‑free framework that replaces costly online action sequence optimization with a goal‑conditioned inverse model. HIJP trains by relabeling future observations as goals, providing action supervision across multiple horizons and shaping visual representations. The method also aligns predicted actions with a latent predictor model to enforce action‑transition consistency, achieving high success rates and significant speedups on visual control tasks.
By Guanchen Du, Jinhao Du, Yanpeng Sun, Shan Zhang, Na Zhao, Zechao Li
arXiv:2610.28483v1 Announce Type: new
Abstract: Learning latent dynamics is fundamental to building action-conditioned world models for planning. Among existing families, Joint Embedding Predictive A...
By Yilei Man, Yang Chen, Jingcai Guo