brihijoshi/granular-similarity-COLING-2020

Code for the paper "The Devil is in the Details: Evaluating Limitations of Transformer-based Methods for Granular Tasks" accepted at COLING 2020

20
/ 100
Experimental

This project helps researchers and data scientists compare text documents for similarity, especially when looking for very specific details rather than just broad topics. It takes in collections of text documents, like news articles or bug reports, and helps evaluate how well different methods can pinpoint fine-grained matches between them. The primary users are those working on natural language processing tasks who need to assess the limitations and strengths of text embedding models for detailed comparison.

No commits in the last 6 months.

Use this if you are a researcher or NLP practitioner evaluating text similarity models and need to understand their performance on tasks requiring granular, detail-oriented matching between documents.

Not ideal if you are looking for an out-of-the-box solution for general, abstract document similarity, or if your primary goal is to find high-level topical matches without needing to evaluate model granularity.

text-similarity natural-language-processing news-deduplication bug-report-analysis text-embeddings
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 4 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

8

Forks

Language

Jupyter Notebook

License

Last pushed

Dec 01, 2020

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/brihijoshi/granular-similarity-COLING-2020"

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