---
title: README
emoji: 📈
colorFrom: yellow
colorTo: red
sdk: static
pinned: true
thumbnail: >-
  https://cdn-uploads.huggingface.co/production/uploads/630461624ec2dfa82a5ad7e7/tGDMlLWfdy8ZxAyqZPzrX.png
---

# 🦾 DAVIAN Robotics

**DAVIAN Robotics** is a research subgroup of the [DAVIAN Lab](https://davian.kaist.ac.kr/) at **KAIST AI Graduate School**, focusing on *Vision-Language-Action (VLA) models* and *robotic manipulation*.  
We aim to build general-purpose embodied intelligence that connects perception, language, and action through scalable data and open research.

Our work spans:
- 🤖 **Embodied AI & Robotic Control** — VLA, imitation learning, diffusion policies  
- 🧠 **Representation & Policy Learning** — multimodal reasoning and cross-domain generalization  
- 🌍 **Open Science** — releasing reproducible code, datasets, and pretrained models  

> “Toward scalable and generalizable embodied intelligence.”