arXiv AI October 7, 2026 By Hojoon Son, Fan Zhang

Seeing the Invisible: Physics-Guided Visual Prompting for Temperature- and Radiation-Aware VLA Navigation

The paper introduces Physics‑Guided Visual Prompting (PG‑VP), a plug‑and‑play module that overlays a virtual obstacle onto the input of a frozen Vision‑Language‑Action model to guide navigation around invisible hazards such as radiation or temperature spikes. PG‑VP performs a physics‑based risk assessment to determine the avoidance direction and dynamically renders the same virtual obstacle across frames, allowing the existing navigation policy to detour without retraining. Experiments on OmniNav with R2R‑CE and RxR‑CE datasets show that PG‑VP steers the policy toward low‑risk actions in 84.9% and 83.2% of cases, while real‑world tests on a robot demonstrate significant safety improvements against thermal and radiation sources.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning

Jun 9

Your Model Already Knows: Attention-Guided Safety Filter for Vision-Language-Action Models

arXiv:2606. 09749v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have demonstrated impressive end-to-end performance across a variety of robotic manipulation tasks.

By Seongbin Park, Fan Zhang, Baharan Mirzasoleiman, Shahriar Talebi, Nader Sehatbakhsh

llms robotics multimodal benchmarks safety

arXiv Machine Learning

Jun 11

Learning What to Say to Your VLA: Mostly Harmless Vision Language Action Model Steering

arXiv:2606. 12299v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models provide a natural language interface to robot control, but the mapping from language to behavior is often brittle and unintuitive: semantically similar instructions can induce drastically different behaviors, while some capabilities may not be elicitable through prompting alone.

By Hyun Joe Jeong, Gokul Swamy, Andrea Bajcsy

llms robotics fine-tuning multimodal

arXiv Computer Vision

Sep 25

Learning to Navigate with Minimal Parameters: Decomposing Visual Navigation Through Closed-Form Geometric Interfaces

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.

By Edward Beng Wai Tan, Siew-Kei Lam

diffusion

arXiv Computer Vision

Sep 17

LEAP: Learning Emergent Active Perception for Quadruped Navigation

arXiv:2609.17628v1 Announce Type: cross Abstract: Active perception allows autonomous agents to select their viewpoints rather than passively process the viewpoints given to them, enabling them to ta...

By \"U. Bora G\"okbakan (WILLOW), St\'ephane Caron (ISIR), Philippe Sou\`eres (LAAS-GEPETTO)

agents robotics

arXiv AI

Oct 1

Blackout vs. Freeze: Analyzing Physical Failure Modes of VLAs under Camera Faults

arXiv:2609.39145v1 Announce Type: cross Abstract: Unreliable visual inputs can harm task performance and cause potential physical safety risks for vision-language-action (VLA) models. We analyze how...

By Heejae Suh, Jongwook Han, Zahra Gholami, Yohan Jo

rag robotics multimodal safety

arXiv AI

Jul 21

Learning Adaptive Safety Margins for Visual Navigation

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.

By Junyi Hu, Shuaihang Yuan, Geeta Chandra Raju Bethala, Anthony Tzes, Yi Fang

diffusion reinforcement-learning robotics safety