File size: 429 Bytes
5e039cd 7317278 342e838 7317278 3b5a7e5 5e039cd 1e1f69c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
---
language:
- en
- zh
license: apache-2.0
tags:
- llava
- vlm
datasets:
- LinkSoul/Chinese-LLaVA-Vision-Instructions
---
The bilingual English/Chinese Baichuan2-7B-Chat VLM trained via LORA for https://arxiv.org/abs/2406.11665.
The Chinese half of the training data used for multimodal alignment and visual instruction tuning is sampled from [here](https://huggingface.co/datasets/LinkSoul/Chinese-LLaVA-Vision-Instructions). |