Hi, thanks for the great work adapting XrayGPT for broader medical imaging and VQA tasks. Are there any plans to release fully trained weights for the MedicalGPT model, especially after fine-tuning on OpenI and ROCO datasets? This would greatly help with reproducibility and further research.
Thanks!