1 Oncodesign Precision Medicine, 18 rue Jean Mazen, Dijon, 21000, France
2Corresponding author: sraieli@oncodesign.com
ArXiv article
This is a technical report showing how is possible to generate an automatic target dossier using LLM, RAG and agents
The advancement of artificial intelligence algorithms has expanded their application to several fields such as the biomedical domain. These new systems, including Large Language Models (LLMs), can be particularly advantageous in drug discovery, which is a very long and expensive process. However, LLMs by themselves lack in-depth knowledge about specific domains and can generate factually incorrect information. Moreover, they are not able to perform more complex actions that imply the usage of external tools. Our work is focused on these two issues. Firstly, we show how the implementation of an advanced RAG system can help the LLM to generate more accurate answers to drug-discovery-related questions. The results show that the answers generated by the LLM with the RAG system surpass in quality the answers produced by the model without RAG. Secondly, we show how to create an automatic target dossier using LLMs and incorporating them with external tools that they can use to execute more intricate tasks to gather data such as accessing databases and executing code. The result is a production-ready target dossier containing the acquired information summarized into a PDF and a PowerPoint presentation.
Chicago formatted citation:
Gabriele Fossi, Youssef Boulaimen, Leila Outemzabet, Nathalie Jeanray, Stephane Gerart, Sebastien Vachenc, Joanna Giemza, and Salvatore Raieli. "SwiftDossier: Tailored Automatic Dossier for Drug Discovery with LLMs and Agents." 2024. arXiv. https://arxiv.org/abs/2409.15817.
BibTeX formatted citation:
@misc{fossi2024swiftdossiertailoredautomaticdossier,
title={SwiftDossier: Tailored Automatic Dossier for Drug Discovery with LLMs and Agents},
author={Gabriele Fossi and Youssef Boulaimen and Leila Outemzabet and Nathalie Jeanray and Stephane Gerart and Sebastien Vachenc and Joanna Giemza and Salvatore Raieli},
year={2024},
eprint={2409.15817},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2409.15817},
}
- a PDF example of an automatically generated target dossier - here
- a PowerPoint generated using a LLM - here
Check the full example.
Table of content:
Example of a pdf page:
Check the full example.
Examples of generated PowerPoint slides:
