Vision-language-action models

Research

FOCA studies future-oriented conditioning for data-efficient VLA adaptation.

Research

Uni-NaVid: A Video-based Vision-Language-Action Model for Unifying Embodied Navigation Tasks.

Research

From Vision-Language-Action Models to a Real-World Robot Learning Stack

Research

VLA-0 provides research code for building vision-language-action models from vision-language models.

Research

Green-VLA: Staged Vision-Language-Action Model for Generalist Robots

Research

VLA-RFT provides reinforcement fine-tuning resources for vision-language-action models.

Research

WholebodyVLA studies latent VLA representations for whole-body locomotion and manipulation.

Research

The official repository of the first version of ACE-Brain foundation model.

Research

Anchor-Align studies VLA fine-tuning with representation anchoring and language-action alignment.

Research

BitVLA implements 1-bit vision-language-action models for robotic manipulation.

Research

CATok studies causal action tokenization for robot policies with conditional annealing in flow matching.

Research

CLP-VLA provides layer-focused fine-tuning and evaluation code for VLA models.

Research

DEAS Isaac GR00T provides setup resources combining DEAS, Isaac GR00T and RoboCasa.

Research

Dexora provides VLA policy resources for high-degree-of-freedom bimanual dexterity.

Research

DynamicVLA provides research resources for dynamic-object robotic manipulation.

Research

EAI2025 Project combines 3D semantic mapping and VLA deployment for manipulation.

Research

Emma-X: An Embodied Multimodal Action Model with Grounded Chain of Thought and Look-ahead Spatial Reasoning

Research

FLUX 3 Action provides world-action modeling and action-prediction resources for robots and simulators.

Research

Google DeepMind robotics models connecting perception and action.

Research

HEX is a whole-body vision-language-action framework for full-sized humanoid robots.

Research

InternVLA-A series studies understanding, generation and action for robotic manipulation.

Research

JanusVLN studies vision-language navigation with separate semantic and spatial memory.

Research

Code of paper "KERV: Kinematic-Rectified Speculative Decoding for Embodied VLA Models" accepted by DAC 2026.

Research

LaST​$_0$​: Latent Spatio-Temporal Chain-of-Thought for Robotic Vision-Language-Action Model