groma_instruct
收藏官方服务：
资源简介：
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](https://arxiv.org/abs/2404.13013) and [github page](https://github.com/FoundationVision/Groma).
提供机构：
maas创建时间：
2025-11-10
