README.md

llms-vote

Code for the Paper "Large Language Models Vote: Prompting for Rare Disease Identification"

The preprint is available on arXiv

Environment

python -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

Reproducing the Dataset

Generating a List of Rare Diseases

Run the following to generate a list of rare diseases using ORDO ontology version 4.2:

mkdir data
./gen_rare_disease.py --ordo data/ORDO_en_4.2.owl --out data/rare_disease.txt

Generating Inverted Index

Get MIMIC-IV first. Then install Rust and run the following:

cd inverted_index
cargo run --release -- ../data/rare_disease.txt ../data/discharge.csv ../data/inverted_index.json

Implementation

llms-vote has been implemented by David Oniani and Jordan Hilsman.