File size: 426 Bytes
73c3414 d9d0008 0a732bf 910fdd1 73c3414 72edfb1 5a8c844 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
---
license: apache-2.0
datasets:
- LinkSoul/Chinese-LLaVA-Vision-Instructions
language:
- en
- zh
tags:
- llava
- vlm
---
The bilingual English/Chinese Llama2-7B-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). |