đ Quilt-LlaVA Model Card
Quilt-LlaVA is an open - source chatbot for medical histopathology research, fine - tuned on specific data sources.
đ Quick Start
This README provides detailed information about the Quilt-LlaVA model, including its details, license, intended use, training and evaluation datasets.
⨠Features
- Medical - Focused: Specifically designed for medical histopathology research.
- Multimodal: Trained on multimodal instruction - following data.
- Open - Source: Based on open - source data and models.
đ Documentation
Model details
Property |
Details |
Model Type |
Quilt-LLaVA is an open - source chatbot trained by fine - tuning LLaMA/Vicuna on histopathology educational video sourced images and GPT - generated multimodal instruction - following data. It is an auto - regressive language model, based on the transformer architecture. |
Citation |
bibtex<br>@article{seyfioglu2023quilt,<br> title={Quilt-LLaVA: Visual Instruction Tuning by Extracting Localized Narratives from Open - Source Histopathology Videos},<br> author={Seyfioglu, Mehmet Saygin and Ikezogwo, Wisdom O and Ghezloo, Fatemeh and Krishna, Ranjay and Shapiro, Linda},<br> journal={arXiv preprint arXiv:2312.04746},<br> year={2023}<br>}<br> |
Model Date |
Quilt-LlaVA-v1.5-7B was trained in November 2023. |
Paper or resources for more information |
https://quilt-llava.github.io/ |
License
Llama 2 is licensed under the LLAMA 2 Community License,
Copyright (c) Meta Platforms, Inc. All Rights Reserved.
Where to send questions or comments about the model:
https://github.com/quilt-llava/quilt-llava.github.io/issues
Intended use
Property |
Details |
Primary intended uses |
The primary use of Quilt-LlaVA is research on medical large multimodal models and chatbots. |
Primary intended users |
The primary intended users of these models are AI researchers. We primarily imagine the model will be used by researchers to better understand the robustness, generalization, and other capabilities, biases, and constraints of large vision - language generative histopathology models. |
Training dataset
- 723K filtered image - text pairs from QUILT - 1M https://quilt1m.github.io/.
- 107K GPT - generated multimodal instruction - following data from QUILT - Instruct https://huggingface.co/datasets/wisdomik/QUILT-LLaVA-Instruct-107K.
Evaluation dataset
A collection of 4 academic VQA histopathology benchmarks
đ License
The model is under the cc - by - nc - 3.0 license.
â ī¸ Important Note
Please read and agree to the following terms: 1. The requester details provided are not faked. 2. The model will not be used for commercial/clinical purposes and will be used for the purpose of scientific research only. 3. The data will not be re - distributed, published, copied, or further disseminated in any way or form whatsoever, whether for profit or not. 4. The right study/paper (Quilt - 1M(https://quilt1m.github.io/) and Quilt - LLaVa (https://quilt - llava.github.io) papers) will be cited in any publication(s) that uses this model/data
đĄ Usage Tip
When using this model, make sure to fill in the extra gated fields accurately, including Email, First and last name, Affiliation, Type of Affiliation, and your intended use. Also, confirm your agreement to the terms of use.