Readme

💡 The official Repo of the paper PROXYQA: An Alternative Framework for Evaluating Long-Form Text Generation with Large Language Models

---

Getting Started

Clone the repo with command:

git clone https://github.com/Namco0816/ProxyQA.git

Loading the proxy-qa with 🤗 Datasets:

from datasets import load_from_disk
proxy_qa = load_from_disk('proxy_qa')
print(proxy_qa)

# Dataset({
#    features: ['meta_question', 'wiki_cover', 'time_sensitive', 'id', 'domain'],
#    num_rows: 100
#})

The distribution of the covered domain

The leaderboard will be available at proxy-qa, currently it is still work in progress.