MiuLab/FactAlign
Source code of our EMNLP 2024 paper "FactAlign: Long-form Factuality Alignment of Large Language Models"
FactAlign helps improve the factual accuracy of long-form text generated by Large Language Models (LLMs). It takes an existing LLM and data about factual accuracy (sentence-by-sentence) to produce a fine-tuned LLM that generates more reliable, factually correct long answers. This is useful for AI engineers or research scientists who are developing or deploying LLMs in applications where factual accuracy is critical.
No commits in the last 6 months.
Use this if you are developing or fine-tuning Large Language Models and need to significantly improve the factual accuracy of their long-form generated content.
Not ideal if you are an end-user simply looking to use an off-the-shelf LLM or if your primary concern is not the factual correctness of long-form outputs.
Stars
19
Forks
1
Language
Jupyter Notebook
License
—
Category
Last pushed
Oct 03, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/MiuLab/FactAlign"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
yueqingliang1/UNBench
Data and code for paper "𝗕𝗲𝗻𝗰𝗵𝗺𝗮𝗿𝗸𝗶𝗻𝗴 𝗟𝗟𝗠𝘀 𝗳𝗼𝗿 𝗣𝗼𝗹𝗶𝘁𝗶𝗰𝗮𝗹 𝗦𝗰𝗶𝗲𝗻𝗰𝗲: 𝗔 𝗨𝗻𝗶𝘁𝗲𝗱 𝗡𝗮𝘁𝗶𝗼𝗻𝘀 𝗣𝗲𝗿𝘀𝗽𝗲𝗰𝘁𝗶𝘃𝗲".
neha13rana/Stereotypical-Bias-Analyzer
In this project, we analyzed biases in ten domains using four datasets and created a useful...
zjunlp/BiasEdit
[TrustNLP@NAACL 2025] BiasEdit: Debiasing Stereotyped Language Models via Model Editing
josmarios/textbias-edu
Code for the article "From hype to evidence: exploring large language models for inter-group...