VLA 線 · Vision-Language-Action
VLA 研究日報VLA 研究日报
VISION-LANGUAGE-ACTION · cs.RO + cs.AI + cs.LG
Vision-Language-Action（VLA）機器人系統 — 整合 cs.RO、cs.AI、cs.LG 三條 arxiv 流。 重點追蹤 flow matching、世界模型、具身推理等前沿方向，每日 09:00 CST 由 Qwen3.5-Plus 自動評級Vision-Language-Action（VLA）机器人系统 — 整合 cs.RO、cs.AI、cs.LG 三条 arxiv 流。 重点追踪 flow matching、世界模型、具身推理等前沿方向，每日 09:00 CST 由 Qwen3.5-Plus 自动评级。
PrefO2O: Zero-Shot Preference-Based Reinforcement Learning via Offline-to-Online Latent Preference Transfer
DRILL: Training World Models to Improve Policies, Not Predict Pixels
From Association to Decision Value: Testing Paired Instruction Dependence for VLA Corpus Selection
T-WLA: Bridging Slow Vision and Fast Touch
ACE-Brain-0.6: Scaling Spatial Grounding for Physical Foundation Models
DUET: Decoupled Upper- and Lower-Body Execution for Task-Agnostic Humanoid Loco-Manipulation
Has GPT-6-Astra Achieved Embodied Intelligence?
Large-scale iontronic skins with neural-hierarchical processing architecture
Human Demonstrations as Post-Training Data for Any Robot Body: Retargeting the Object, Not the Hand
ACCRUE: Continual Off-Policy Reinforcement Learning for Robot Foundation Models
PanTact: Teaching Whole-Body Dexterity with a Wearable Contact Interface
HiPHI: A Large-Scale Benchmark for High-Precision Human Motion and Object Interaction
Splat2Act: From Captured Scenes to Personalized VLA Manipulation
Gamma: Grounded Multi-Agent Memory with Mechanical Audit for Memory-Intensive Manipulation
When Listening Becomes Easier: Scrubbing Visual Cues for Shortcut-Free VLAs
Co-Evolving Robot Orchestrators and Policies through Deployment
Physics filtering favors the generalization of robot learning
Physics filtering favors the generalization of robot learning
Physics filtering favors the generalization of robot learning
Physics filtering favors the generalization of robot learning
Fiatlux: A Long-Horizon Benchmark for Humanoid Ladder Climbing and Light-Bulb Replacement
FineART: Fine-grained Annotated Robotic Trajectory Dataset and Vision-Language-Action Model for Bimanual Manipulation
CoHuB: A Simulation Benchmark for Multi-Humanoid Collaboration
Policy-Calibrated DAgger: Offline Calibrated Noise Injection for Imitation Learning
Morphometric Imitation: From Morphology and Contact Aware Hand Retargeting to Sim-to-Real Visuomotor Policy
InternW0: A Foundational Physical World Model for Efficient Real-World Interactions
LIBERO-VPro: Benchmarking Closed-Loop Visual Robustness of Robotic Foundation Models
LIBERO-VPro: Benchmarking Closed-Loop Visual Robustness of Robotic Foundation Models
Shake to Learn: Dynamic Interrogation of Hidden Object Physics for Robotic Manipulation with Physical Reservoir Computing
Beyond Patch Removal: Persistent Adversarial Effects in Vision-Language-Action Policies
Energy-Regularized Imitation Learning for Force- and Work-Aware Robotic Manipulation
AthenaZero: A low-inertia, bimanual robot for dynamic manipulation
ShieldVLA: Feasibility-Aware Safety Alignment for Vision-Language-Action Models
Marker Information Efficacy: guiding marker design through deformation information encoding in vision-based tactile sensing
World in World: Explore the World with World Models
An FBG tactile sensor array and self-supervised contrastive learning transformer for tumor depth estimation in robotic palpation
Robots Are Learning to Feel
Physics filtering favors the generalization of robot learning
Physics filtering favors the generalization of robot learning
Physics filtering favors the generalization of robot learning
Physics filtering favors the generalization of robot learning
Physics filtering favors the generalization of robot learning