Cover Image

Self-Improving Vision-Language-Action Models with Data Generation via Residual RL

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Xiao, Wenli, Lin, Haotian, Peng, Andy, Xue, Haoru, He, Tairan, Xie, Yuqi, Hu, Fengyuan, Wu, Jimmy, Luo, Zhengyi, Fan, Linxi "Jim", Shi, Guanya, Zhu, Yuke
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition Robotics
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

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

VIRAL: Visual Sim-to-Real at Scale for Humanoid Loco-Manipulation
by: He, Tairan, et al.
Published: (2025)

HOVER: Versatile Neural Whole-Body Controller for Humanoid Robots
by: He, Tairan, et al.
Published: (2024)

Learning Human-to-Humanoid Real-Time Whole-Body Teleoperation
by: He, Tairan, et al.
Published: (2024)

WoCoCo: Learning Whole-Body Humanoid Control with Sequential Contacts
by: Zhang, Chong, et al.
Published: (2024)

Safe Deep Policy Adaptation
by: Xiao, Wenli, et al.
Published: (2023)

CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation
by: Fu, Max, et al.
Published: (2026)

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

AnyCar to Anywhere: Learning Universal Dynamics Model for Agile and Adaptive Mobility
by: Xiao, Wenli, et al.
Published: (2024)

Agile Mobility with Rapid Online Adaptation via Meta-learning and Uncertainty-aware MPPI
by: Kalaria, Dvij, et al.
Published: (2024)

OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning
by: He, Tairan, et al.
Published: (2024)

ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills
by: He, Tairan, et al.
Published: (2025)

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

SCIZOR: A Self-Supervised Approach to Data Curation for Large-Scale Imitation Learning
by: Zhang, Yu, et al.
Published: (2025)

Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealing
by: Xue, Haoru, et al.
Published: (2024)

HDMI: Learning Interactive Humanoid Whole-Body Control from Human Videos
by: Weng, Haoyang, et al.
Published: (2025)

Bridging Adaptivity and Safety: Learning Agile Collision-Free Locomotion Across Varied Physics
by: Zhong, Yichao, et al.
Published: (2025)

Residual Off-Policy RL for Finetuning Behavior Cloning Policies
by: Ankile, Lars, et al.
Published: (2025)

AMAGO: Scalable In-Context Reinforcement Learning for Adaptive Agents
by: Grigsby, Jake, et al.
Published: (2023)

Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on Humanoids
by: Lin, Toru, et al.
Published: (2025)

TD-M(PC)$^2$: Improving Temporal Difference MPC Through Policy Constraint
by: Lin, Haotian, et al.
Published: (2025)

Hold My Beer: Learning Gentle Humanoid Locomotion and End-Effector Stabilization Control
by: Li, Yitang, et al.
Published: (2025)

Sampling-Based System Identification with Active Exploration for Legged Robot Sim2Real Learning
by: Sobanbabu, Nikhil, et al.
Published: (2025)

LeVERB: Humanoid Whole-Body Control with Latent Vision-Language Instruction
by: Xue, Haoru, et al.
Published: (2025)

CHIP: Adaptive Compliance for Humanoid Control through Hindsight Perturbation
by: Chen, Sirui, et al.
Published: (2025)

DexMimicGen: Automated Data Generation for Bimanual Dexterous Manipulation via Imitation Learning
by: Jiang, Zhenyu, et al.
Published: (2024)

SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control
by: Luo, Zhengyi, et al.
Published: (2025)

Atomicity for Agents: Exposing, Exploiting, and Mitigating TOCTOU Vulnerabilities in Browser-Use Agents
by: Jiang, Linxi, et al.
Published: (2026)

DEAS: DEtached value learning with Action Sequence for Scalable Offline RL
by: Kim, Changyeon, et al.
Published: (2025)

RAMBO: RL-Augmented Model-Based Whole-Body Control for Loco-Manipulation
by: Cheng, Jin, et al.
Published: (2025)

Self-Supervised Meta-Learning for All-Layer DNN-Based Adaptive Control with Stability Guarantees
by: He, Guanqi, et al.
Published: (2024)

ResWM: Residual-Action World Model for Visual RL
by: Zhang, Jseen, et al.
Published: (2026)

HumanoidMimicGen: Data Generation for Loco-Manipulation via Whole-Body Planning
by: Lin, Kevin, et al.
Published: (2026)

EgoScale: Scaling Dexterous Manipulation with Diverse Egocentric Human Data
by: Zheng, Ruijie, et al.
Published: (2026)

NitroGen: An Open Foundation Model for Generalist Gaming Agents
by: Magne, Loïc, et al.
Published: (2026)

World Action Models are Zero-shot Policies
by: Ye, Seonghyeon, et al.
Published: (2026)

RL Token: Bootstrapping Online RL with Vision-Language-Action Models
by: Xu, Charles, et al.
Published: (2026)

Sim-and-Real Co-Training: A Simple Recipe for Vision-Based Robotic Manipulation
by: Maddukuri, Abhiram, et al.
Published: (2025)

$π_\texttt{RL}$: Online RL Fine-tuning for Flow-based Vision-Language-Action Models
by: Chen, Kang, et al.
Published: (2025)

QLIP: Text-Aligned Visual Tokenization Unifies Auto-Regressive Multimodal Understanding and Generation
by: Zhao, Yue, et al.
Published: (2025)