Adaptive Vision-Language Grasping via Composable Foundation Priors and Generalizable Grasp Synthesis
cs.RO, cs.AI, cs.CV
Submitted: 2026-09-03
Updated: 2026-09-30
Code: https://github.com/SeungBack/gc6d-contact-graspnet
Project page: https://adarobovlg.github.io
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- DINOv2: Learning Robust Visual Features without Supervision
- DINOv3
- RoboBrain: A Unified Brain Model for Robotic Manipulation from Abstract to Concrete
- RoboBrain 2.0 Technical Report
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- $\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization
- GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data
- DexGraspVLA: A Vision-Language-Action Framework Towards General Dexterous Grasping
- DexVLG: Dexterous Vision-Language-Grasp Model at Scale
- Being-H0: Vision-Language-Action Pretraining from Large-Scale Human Videos
- OmniDexVLG: Learning Dexterous Grasp Generation from Vision Language Model-Guided Grasp Semantics, Taxonomy and Functional Affordance
- AutoDex: An Automated Real-World System for Dexterous Grasping Data Collection
- SAM 3: Segment Anything with Concepts
- GraspClutter6D: A Large-scale Real-world Dataset for Robust Perception and Grasping in Cluttered Scenes
- RoboPoint: A Vision-Language Model for Spatial Affordance Prediction for Robotics
- GraspSplats: Efficient Manipulation with 3D Feature Splatting
- ThinkGrasp: A Vision-Language System for Strategic Part Grasping in Clutter
- Lan-grasp: Using Large Language Models for Semantic Object Grasping and Placement
- ClutterDexGrasp: A Sim-to-Real System for General Dexterous Grasping in Cluttered Scenes
- CEDex: Cross-Embodiment Dexterous Grasp Generation at Scale from Human-like Contact Representations
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
- Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving