Particle.news
Science ❯ Computer Science ❯ Artificial Intelligence
Vision-Language-Action Models Human-Robot Interaction Decision Making Manipulation Tasks Autonomous Systems Action Prediction Embodied Intelligence Embodied AI Control Systems Policy Learning Robotic Manipulation Benchmarking Autonomous Driving Action Generation Safety in AI Embodied Agents Reinforcement Learning Task Automation Action Recognition Spatial Reasoning Vision-Language Integration Action Models Interdisciplinary Studies Natural Language Processing Self-Improvement Cognitive Robotics Facial Recognition AI Engineering 3D Modeling Human-Computer Interaction Metrics for Interaction Evaluation Service Robots Simulation Physical AI Planning Algorithms Security Action Planning Control Theory Task Planning Manipulation Planning Tactile Feedback Systems Action Modeling Multimodal Learning Motion Reasoning Data Efficiency Vision-Language-Action Systems Task Progress Estimation Motion Planning Spatiotemporal Manipulation Spatial Intelligence Vision-Language Models Multimodal Reasoning Navigation Systems Multimodal Systems Physical Constraints End-Effector Geometry OOD Execution Action Representation Model Adaptation Data-Centric Approaches Vision-Language Processing Sports Technology Continual Learning Demonstration Techniques Sensory Processing Vision Systems Behavior Representation Operational Safety Policy Optimization Techniques Geometric Representation Evaluation Methods Task Benchmarking Vision-language-action Policies Task Conditioning Task Performance Improvement Safety in Robotics Action Flow Systems Collision Avoidance Future Prediction System Identification Instruction Following Multi-Robot Coordination Performance Evaluation Task Optimization Task Execution 3D Manipulation Action Generation Techniques Embodied Learning Task-Aware Decision Making Human-Robot Collaboration Biomedical Applications Model Evaluation Imitation Learning Manipulation Techniques Action Control Systems Semantic Grounding Inference Methods Action Normalization Flow Policy Optimization
A cluster of July preprints retools model inputs, action interfaces, evaluation methods, runtimes to make pretrained vision‑language backbones reliable for real‑robot control.
