arXiv:2609.38383v1 Announce Type: cross Abstract: Random exploration reveals how an environment can be traversed before a goal is specified. Can this experience support long-range planning without po...
arXiv:2606. 08992v1 Announce Type: cross Abstract: Vision-and-Language Navigation in continuous environments requires agents to understand the spatial structure of previously unseen environments in order to follow language instructions.
Conventional visual navigation policies often struggle with myopic decision-making and mode collapse in complex environments. While world models offer a promising alternative, existing paradigms typically isolate perception, generation, and control, failing to capture their shared spatio-temporal dynamics.
The paper introduces EvolvingNav, a system that builds a time‑indexed belief about moving targets in dynamic environments by combining timestamped 3D object histories with a persistence‑relocation model. It uses an event‑driven filter to update beliefs over time, incorporates RGB‑D evidence, and applies a zero‑shot vision‑language controller for action selection. The authors also present EvoWorld‑Bench, a large benchmark of human‑activity‑based scenes, and demonstrate that EvolvingNav outperforms baselines in both simulation and real‑robot experiments, especially when temporal patterns are learnable.
SearchWorld introduces a recurrent state‑space world model that couples explicit spatial memory with value‑guided imagination for UAV object search. The model maintains bird’s‑eye‑view exploration and obstacle memory, decodes a task‑aware spatial value layer, and uses a cognition‑action network to improve policy through imagined rollouts without a separate critic. Experiments on UAV‑ON show a 23.8% success rate, surpassing the best published agent (19.5%) and achieving 35.5% oracle success while remaining robust on unseen scenes.
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.
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.
LT-Mem introduces a volatility‑aware memory evolution framework for lifelong scene understanding, combining spatially aligned instance‑level 3D perception with temporal reasoning. It uses a multi‑session SLAM backbone, a reasoning layer that scores evidence and selects memory actions, and a Tri‑Memory structure (Live, Delta, Meta) to preserve current states and event histories. The accompanying LT‑VQA dataset provides multi‑session recordings, persistent identity annotations, and temporal QA pairs, and experiments show LT‑Mem outperforms baselines while using far fewer tokens.
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...
arXiv:2606.20627v2 Announce Type: replace Abstract: Joint-Embedding Predictive Architectures (JEPAs) enable agents to plan in latent space by imagining the outcomes of candidate actions, yet task spe...
arXiv:2606. 18888v1 Announce Type: new Abstract: Navigation in partially observable environments presents a significant challenge for autonomous agents, requiring effective decision-making with limited sensory information in unknown environments.
Learning latent dynamics is fundamental to building action-conditioned world models for planning. Among existing families, Joint Embedding Predictive Architectures (JEPAs) offer a natural framework by...