PhD Student in Speech Processing Applied to Medical Domain at Laboratoire Informatique d'Avignon / Avignon University / Zenidoc
Avignon
Joined September 2021
- Tweets34
- Following32
- Followers40
- Likes103
labrak yanis retweeted
How I Built a Medical RAG Chatbot Using BioMistral|Langchain | FREE Colab|ALL OPENSOURCE #ai
youtu.be/E53hc-jcUeE
labrak yanis retweeted
BioMistral is the new open-source model for medical domains. It has a completely free license, without concern for royalties.
Use it yourself - with no-code needed - using the guide below.
We are entering an era of domain-specialized, free models.
labrak yanis retweeted
Yeah.
BioMistral seems pretty cool too:
huggingface.co/BioMistral/Bi…
labrak yanis retweeted
#LLM BioMistral is a new 7B foundation model for medical domains, based on Mistral and further trained PubMed Central. - top open-source medical Large Language Model (LLM) in its weight class - Apache License - huggingface.co/BioMistral/Bi…
labrak yanis retweeted
BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains - arxiv.org/abs/2402.10373 🧬🖥️ huggingface.co/BioMistral
labrak yanis retweeted
New model added to the leaderboard!
Model Name
hf.co/BioMistral/BioMistral-…
Overall rank: 2697
Rank in 7B category: 1190
Benchmarks
Average: 52.33
ARC: 54.27
HellaSwag: 79.09
MMLU: 55.56
TruthfulQA: 51.61
Winogrande: 73.48
GSM8K: 0.0
labrak yanis retweeted
New model added to the leaderboard!
Model Name
hf.co/BioMistral/BioMistral-…
Overall rank: 2051
Rank in 7B category: 1053
Benchmarks
Average: 57.03
ARC: 58.28
HellaSwag: 79.87
MMLU: 57.34
TruthfulQA: 55.61
Winogrande: 76.09
GSM8K: 15.01
labrak yanis retweeted
New model added to the leaderboard!
Model Name
hf.co/BioMistral/BioMistral-…
Overall rank: 2622
Rank in 7B category: 1163
Benchmarks
Average: 52.91
ARC: 55.46
HellaSwag: 79.59
MMLU: 56.29
TruthfulQA: 52.2
Winogrande: 73.72
GSM8K: 0.23
labrak yanis retweeted
BioMistral: A Collection of Open-Source Pretrained Large Language Models for Medical Domains. arxiv.org/abs/2402.10373
labrak yanis retweeted
This development is not just about advancing healthcare through AI but also democratizing access to cutting-edge technology. With BioMistral's release, the potential for enhancing medical research, diagnostics, and patient care is immense, underscoring the critical role of OS.
labrak yanis retweeted
This model's multilingual capabilities further extend its reach, promising to bridge language barriers in medical knowledge dissemination. As we embrace models like BioMistral, the future of healthcare looks not just more intelligent but more inclusive.
#HealthcareInnovation
labrak yanis retweeted
BioMistral基于Mistral微调的医学领域 7B 模型，包括基本模型、微调和量化版本。
BioMistral is a new 7B foundation model for medical domains, based on Mistral and further trained PubMed Central.
- top open-source medical Large Language Model (LLM) in its weight class
- Apache License
- includes base models, fine tunes, and quantized versions.
ALT Screenshot of the model card on the Hub.
labrak yanis retweeted
@huggingface org: huggingface.co/BioMistral
Base model: huggingface.co/BioMistral/Bi…
Paper: huggingface.co/papers/2402.1…
labrak yanis retweeted
BioMistral buff.ly/48sqGzv just came out last week. It is an open-source multilingual Large Language Model (LLM) specifically designed for the medical domain. It is based on the robust foundation of Mistral and further pre-trained on PubMed Central.
#AI #OpenSource
labrak yanis retweeted
Very interesting for those in the medical space. Models also available on the Hugging Face Hub.
HF Page: huggingface.co/BioMistral
Paper: arxiv.org/abs/2402.10373
labrak yanis retweeted
BioMistral 7B just arrived!
A Multilingual Medical LLM for the medical domain. BioMistral's multilingual potential and robustness across diverse linguistic contexts, with a focus on truthfulness and calibration in multilingual environments. The 7B model surpasses all competitive open-source medical LLMs within the same 7B category on a benchmark consisting of 10 medical question answering tasks in English and automatically translated into 7 additional languages (French, Italian, German, Chinese, Arabic, Spanish, and Portuguese).
Check out the:
- BioMistral 7B : huggingface.co/BioMistral/Bi…
- BioMistral 7B DARE 🍄 : huggingface.co/collections/B…
- AWQ Quantized models (less than 4.68 GB of VRAM with not performance degradation) : huggingface.co/collections/B…
- 📊 Benchmark : huggingface.co/datasets/BioM…
- Paper: arxiv.org/pdf/2402.10373.pdf
Let's give the contributors to the open source sphere a big 👏 @LabrakYanis @r_dufour @AdrienBazoge @CNRSinformatics @LaboLS2N @taln_ls2n
#BioMistral #MedicalLLM #Multilingual #NLP
🚀 🏥 Very proud to announce BioMistral, a collection of open-source pre-trained LLMs for the medical domain
📰Arxiv: tinyurl.com/3xk8hua6
🏥 BioMistral 7B model: tinyurl.com/mubkfprp
More info: tinyurl.com/5etvbvkz
@CNRSinformatics @LaboLS2N @taln_ls2n @LabrakYanis
labrak yanis retweeted
🚀 🏥 Very proud to announce BioMistral, a collection of open-source pre-trained LLMs for the medical domain
📰Arxiv: tinyurl.com/3xk8hua6
🏥 BioMistral 7B model: tinyurl.com/mubkfprp
More info: tinyurl.com/5etvbvkz
@CNRSinformatics @LaboLS2N @taln_ls2n @LabrakYanis
labrak yanis retweeted
BioMistral is a new 7B foundation model for medical domains, based on Mistral and further trained PubMed Central.
- top open-source medical Large Language Model (LLM) in its weight class
- Apache License
- includes base models, fine tunes, and quantized versions.
ALT Screenshot of the model card on the Hub.
Submitted only 4 months after the beginning of my PhD, "DrBERT: A Robust Pre-trained Model in French for Biomedical and Clinical domains" is my 1st paper accepted #ACL2023 main conference! Scripts & Models are available at drbert.univ-avignon.fr/.
arXiv: arxiv.org/abs/2304.00958