Vintern-1B-v2-ViTable-docvqa

Report Link👁️

Vintern-1B-v2-ViTable-docvqa is a fine-tuned version of the 5CD-AI/Vintern-1B-v2 multimodal model for the Vietnamese DocVQA (Table data)

Benchmarks

Model	ANLS	Semantic Similarity	MLLM-as-judge (Gemini)
Gemini 1.5 Flash	0.35	0.56	0.40
Vintern-1B-v2	0.04	0.45	0.50
Vintern-1B-v2-ViTable-docvqa	0.50	0.71	0.59

Usage

Check out this 🤗 HF Demo, or you can open it in Colab:

Citation:

@misc{doan2024vintern1befficientmultimodallarge,
      title={Vintern-1B: An Efficient Multimodal Large Language Model for Vietnamese}, 
      author={Khang T. Doan and Bao G. Huynh and Dung T. Hoang and Thuc D. Pham and Nhat H. Pham and Quan T. M. Nguyen and Bang Q. Vo and Suong N. Hoang},
      year={2024},
      eprint={2408.12480},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2408.12480}, 
}