Cover Image

When Backdoors Meet Partial Observability: Attacking Real-World Reinforcement Learning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Huang, Tairan, Ye, Qingqing, Jin, Yulin, Lian, Jiawei, Xiao, Yaxin, Wang, Yi, Hu, Haibo
Format:	Preprint
Published:	2026
Subjects:	Robotics Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning
by: Huang, Tairan, et al.
Published: (2025)

Equivariant Reinforcement Learning under Partial Observability
by: Nguyen, Hai, et al.
Published: (2024)

Emergent Active Perception and Dexterity of Simulated Humanoids from Visual Reinforcement Learning
by: Luo, Zhengyi, et al.
Published: (2025)

Reinforcement Learning Meets Visual Odometry
by: Messikommer, Nico, et al.
Published: (2024)

Sim2Real-AD: A Modular Sim-to-Real Framework for Deploying VLM-Guided Reinforcement Learning in Real-World Autonomous Driving
by: Huang, Zilin, et al.
Published: (2026)

Adaptive Compensation for Robotic Joint Failures Using Partially Observable Reinforcement Learning
by: Pham, Tan-Hanh, et al.
Published: (2024)

Autonomous Embodied Agents: When Robotics Meets Deep Learning Reasoning
by: Bigazzi, Roberto
Published: (2025)

Guided Diffusion-based Generation of Adversarial Objects for Real-World Monocular Depth Estimation Attacks
by: Chen, Yongtao, et al.
Published: (2025)

SimScale: Learning to Drive via Real-World Simulation at Scale
by: Tian, Haochen, et al.
Published: (2025)

When Attention Betrays: Erasing Backdoor Attacks in Robotic Policies by Reconstructing Visual Tokens
by: Li, Xuetao, et al.
Published: (2026)

RoboTron-Sim: Improving Real-World Driving via Simulated Hard-Case
by: Xiao, Baihui, et al.
Published: (2025)

Opening the Sim-to-Real Door for Humanoid Pixel-to-Action Policy Transfer
by: Xue, Haoru, et al.
Published: (2025)

Synchronous vs Asynchronous Reinforcement Learning in a Real World Robot
by: Parsaee, Ali, et al.
Published: (2025)

Scene Informer: Anchor-based Occlusion Inference and Trajectory Prediction in Partially Observable Environments
by: Lange, Bernard, et al.
Published: (2023)

Collision-Aware Object-Goal Visual Navigation via Two-Stage Deep Reinforcement Learning
by: Wang, Hongwu, et al.
Published: (2025)

WorldEval: World Model as Real-World Robot Policies Evaluator
by: Li, Yaxuan, et al.
Published: (2025)

Driving in Corner Case: A Real-World Adversarial Closed-Loop Evaluation Platform for End-to-End Autonomous Driving
by: Geng, Jiaheng, et al.
Published: (2025)

Class-feature Watermark: A Resilient Black-box Watermark Against Model Extraction Attacks
by: Xiao, Yaxin, et al.
Published: (2025)

Tracking Tumors under Deformation from Partial Point Clouds using Occupancy Networks
by: Henrich, Pit, et al.
Published: (2024)

UniPrototype: Humn-Robot Skill Learning with Uniform Prototypes
by: Hu, Xiao, et al.
Published: (2025)

Dexterous Grasping with Real-World Robotic Reinforcement Learning
by: Huang, Dongchi, et al.
Published: (2025)

Heterogeneous Team Coordination on Partially Observable Graphs with Realistic Communication
by: Zhou, Yanlin, et al.
Published: (2024)

Dreamitate: Real-World Visuomotor Policy Learning via Video Generation
by: Liang, Junbang, et al.
Published: (2024)

Learning Camera Movement Control from Real-World Drone Videos
by: Hou, Yunzhong, et al.
Published: (2024)

Online Gaussian Process State-Space Model: Learning and Planning for Partially Observable Dynamical Systems
by: Park, Soon-Seo, et al.
Published: (2019)

Tactile-based Reinforcement Learning for Adaptive Grasping under Observation Uncertainties
by: Hu, Xiao, et al.
Published: (2025)

Beware Untrusted Simulators -- Reward-Free Backdoor Attacks in Reinforcement Learning
by: Rathbun, Ethan, et al.
Published: (2026)

DiMSam: Diffusion Models as Samplers for Task and Motion Planning under Partial Observability
by: Fang, Xiaolin, et al.
Published: (2023)

General Place Recognition Survey: Towards Real-World Autonomy
by: Yin, Peng, et al.
Published: (2024)

On Mobile Ad Hoc Networks for Coverage of Partially Observable Worlds
by: Meriaux, Edwin, et al.
Published: (2025)

Multi-Visual-Inertial System: Analysis, Calibration and Estimation
by: Yang, Yulin, et al.
Published: (2023)

Ego to World: Collaborative Spatial Reasoning in Embodied Systems via Reinforcement Learning
by: Zhou, Heng, et al.
Published: (2026)

Manipulate-Anything: Automating Real-World Robots using Vision-Language Models
by: Duan, Jiafei, et al.
Published: (2024)

EmbodieDreamer: Advancing Real2Sim2Real Transfer for Policy Training via Embodied World Modeling
by: Wang, Boyuan, et al.
Published: (2025)

EARL: Towards a Unified Analysis-Guided Reinforcement Learning Framework for Egocentric Interaction Reasoning and Pixel Grounding
by: Su, Yuejiao, et al.
Published: (2026)

FlowHijack: A Dynamics-Aware Backdoor Attack on Flow-Matching Vision-Language-Action Models
by: An, Xinyuan, et al.
Published: (2026)

Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion
by: He, Tairan, et al.
Published: (2024)

Persistent Robot World Models: Stabilizing Multi-Step Rollouts via Reinforcement Learning
by: Bardhan, Jai, et al.
Published: (2026)

VR-Robo: A Real-to-Sim-to-Real Framework for Visual Robot Navigation and Locomotion
by: Zhu, Shaoting, et al.
Published: (2025)

Efficient Surgical Robotic Instrument Pose Reconstruction in Real World Conditions Using Unified Feature Detection
by: Liang, Zekai, et al.
Published: (2025)