Colbert Open-Source Passage Search Model - Free Deployment for Efficient Passage Search Functionality

Colbert

Developed by webis

This model is a BERT-based ColBERT model designed for efficient passage retrieval through contextual late interaction.

Downloads 1,242

Release Time : 8/28/2024

Model Overview

Lightning IR ColBERT is a BERT-based dual-encoder model specifically designed for information retrieval tasks, enabling efficient passage retrieval.

Efficient Passage Retrieval

Achieves efficient passage retrieval through BERT-based contextual late interaction.

Dual Encoder Architecture

Employs a dual-encoder architecture capable of simultaneously encoding queries and documents.

Fine-tuning Support

Supports fine-tuning via Lightning IR to adapt to various information retrieval tasks.

Text Ranking

Passage Retrieval

Information Retrieval

Information Retrieval

MS MARCO Passage Retrieval

Used for passage retrieval tasks on the MS MARCO dataset.

TREC Deep Learning Evaluation

Evaluated on the TREC Deep Learning 2019 and 2020 datasets.

Property	Details
Pipeline Tag	Text Ranking
Library Name	Lightning IR
Base Model	google-bert/bert-base-uncased
Tags	bi-encoder
License	Apache-2.0

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base