The paper introduces Visual Learned Distance (VLD), a framework that separates perception from policy learning for image-goal navigation. VLD trains a self-supervised distance-to-goal predictor on thousands of hours of video, then uses a reinforcement learning policy trained in simulation with a learned noise model (GeoNoise) to mimic the predictor’s errors. At deployment, the policy uses VLD predictions zero‑shot, achieving high success rates in simulated environments and on a physical robot without retraining.
arXiv:2606. 17386v1 Announce Type: cross Abstract: End-to-end autonomous driving has achieved state-of-the-art performance on benchmarks and real-world deployments.
arXiv:2602. 13977v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) promises to unlock capabilities beyond imitation learning for Vision--Language--Action (VLA) models, but its requirement for massive real-world interaction prevents direct deployment on physical robots.
The paper introduces a compact visual navigation system that decomposes the task into three analytically‑computed geometric interfaces and three small learned modules: an egress predictor, a navigation predictor, and an endpoint‑pinned residual diffusion generator. Only 0.58 M of the 23 M parameters are trained on 44 k frames, achieving competitive success rates and the lowest collision rate among evaluated methods across 6 060 point‑goal episodes in 60 environments. The design allows further parameter reduction by replacing the frozen image encoder with a 0.54 M MobileNetV2, supports zero‑shot deployment on a Jetson Orin Nano UGV, and enables transparent failure analysis under sensor corruption.
arXiv:2603. 25937v2 Announce Type: replace-cross Abstract: Visual Navigation Models (VNMs) promise generalizable, robot navigation by learning from large-scale visual demonstrations.
arXiv:2607. 20785v1 Announce Type: cross Abstract: Deploying navigation systems at scale requires a recipe that minimizes sensor assumptions, generalizes across robot embodiments, and trains efficiently.
SimForcing is a framework that distills motion priors from simulation into real‑domain robot world models. It transfers motion knowledge via latent‑motion distillation and uses multi‑block simulation conditioning with dropout to guide predictions without relying on accurate simulation outputs. The jointly trained student produces both simulation conditions and real‑domain videos, achieving state‑of‑the‑art metrics on Bridge and demonstrating applicability across robot datasets, while also improving downstream policy learning when used to initialize vision‑language‑action models.
End-to-end autonomous driving has achieved state-of-the-art performance on benchmarks and real-world deployments. Its standard training recipe, however, is expensive across all existing paradigms: imi...
arXiv:2607. 18200v1 Announce Type: cross Abstract: Robots in cluttered indoor spaces often fail not because they cannot generate collision-free paths, but because a fixed safety margin is mis-calibrated: conservative margins cause detours and timeouts, while permissive margins lead to near-boundary shortcuts under perception bias.
The paper introduces a reinforcement learning post‑training scheme that trains robot world models on their own autoregressive rollouts, using a contrastive RL objective adapted from diffusion models. It also proposes a training protocol that compares multiple variable‑length futures, a multi‑view visual fidelity reward, and demonstrates state‑of‑the‑art rollout fidelity on the DROID dataset, outperforming baselines on LPIPS, SSIM, and human preference tests.
TerraTransfer++ demonstrates a demonstration‑free, end‑to‑end autonomous driving framework that leverages self‑play in vectorized simulators to pretrain a policy. The method distills self‑play rollouts into a flow‑matching trajectory teacher and trains lightweight adapters on a frozen vision backbone, using action KL divergence, a batch‑relational low‑rank loss, and trajectory‑matching loss. The resulting closed‑loop policy surpasses all published imitation‑based methods on the HUGSim benchmark, while the planner achieves state‑of‑the‑art performance on NAVSIM navhard and navtest without any human demonstrations.
arXiv:2610. 29845v1 Announce Type: cross Abstract: Autonomous mobile robots operating in dense crowds and cluttered sites require $360^\circ$ perception to react to surrounding hazards.
arXiv:2610.16558v1 Announce Type: new Abstract: World models capture environment dynamics and support counterfactual reasoning for planning. In image-goal navigation, where the target often lies beyo...