Vision-Language-Action

2026

VLA Post-Training

Part I: a controlled comparison of LoRA, QLoRA, and DoRA for post-training GR00T N1.7 on LIBERO-10 Long.

Open project

2026

VLA Robot Learning

Part II: why multi-robot data is difficult to pool, and how open generalist policies adapt across embodiments.

Open project

2026

VLA Robot Learning

Part I: the policy foundations beneath VLAs—from behavior cloning to action chunks, diffusion, and flow matching.

Open project

2026

Vision-language-action post-training and sim-to-real robotic manipulation

Worked on post-training and evaluation workflows for vision-language-action models, with a focus on robotic manipulation and transfer from simulation to real systems.

Open project