Hi, thanks for the really awesome open-source efforts, I wonder whether you plan to release a V-L-A model that is trained/finetuned on the dataset.
Hi, thanks for the amazing open-source work. I was wondering whether you plan to release a VLA model trained or fine-tuned on this dataset. In my view, the most useful data is data that can actually train a capable model, so a released model would be a very valuable way to demonstrate the dataset’s effectiveness.
Thanks for taking interest in the dataset. Yes, we are working on a VLA fine-tuned model / hand trajectory prediction model on the dataset to prove the efficiency of the data. We hope to make an official release on this soon.
As we work towards it, we also welcome community participation, if someone is interested. If someone is willing to contribute to our open source ecosystem by contributing training code and train a model, they can submit the code with a training plan in our github repository - we might be able to support with compute credits.