OpenMatch/COCO-DR

[EMNLP 2022] This is the code repo for our EMNLP‘22 paper "COCO-DR: Combating Distribution Shifts in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learning".

33
/ 100
Emerging

This project helps improve the accuracy of searching for relevant documents or passages, even when the query data is very different from the data used to train the search model. It takes a collection of documents and queries, and outputs a more robust search model that can find better matches across various types of text. This is designed for anyone building advanced search or recommendation systems who needs precise results without extensive fine-tuning.

No commits in the last 6 months.

Use this if you need to build a text search system that performs well on diverse or unexpected document collections without needing to retrain the model for each new domain.

Not ideal if your search needs are basic and don't involve significant shifts in the type of text being queried or searched.

information-retrieval zero-shot-learning document-search text-matching domain-adaptation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

50

Forks

4

Language

Python

License

MIT

Last pushed

Oct 12, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/OpenMatch/COCO-DR"

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