psunlpgroup/FairSumm

Code and Data for NAACL 2024 paper: Fair Abstractive Summarization of Diverse Perspectives

27
/ 100
Experimental

This project helps you create summaries of many individual comments or reviews, like those found in product reviews or political discussions, to ensure all viewpoints are fairly represented. You provide a collection of user-generated texts with diverse perspectives (e.g., positive/negative reviews, different political stances) and it generates a summary that accurately reflects all these different groups. This is for anyone who needs to summarize large volumes of feedback or opinions and wants to avoid unintentionally biased summaries.

No commits in the last 6 months.

Use this if you need to generate summaries of user-generated content, such as product reviews or public comments, and are concerned that automated summaries might overlook or underrepresent certain opinions or groups.

Not ideal if your summarization task involves factual articles, news reports, or documents where the primary goal is factual extraction rather than balancing diverse, subjective viewpoints.

customer-feedback market-research public-opinion content-analysis policy-review
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 6 / 25

How are scores calculated?

Stars

12

Forks

1

Language

Jupyter Notebook

License

Apache-2.0

Last pushed

Mar 30, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/psunlpgroup/FairSumm"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.