Groma Instruct contains 30k visually grounded conversations for instruction finetuning, which is the first grounded chat dataset constructed with both visual and textual prompts, leveraging the powerful GPT-4V for data generation.
For more information about this dataset and its usage, please take a look at our paper and github page.
