This post is a brief summary about the paper that I read for my study and curiosity, so I shortly arrange the content of the paper titled ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT. Khattab et al., arXiv 2020 that I read and studied.

They propose effective ranking models based on Language model, Rather the ranking models based on Language model increase the computation cost, in particular, as they muse deal with each query-document pair to obtain relevance score.

To address this, they present ColBERT, which is a novel ranking model that adapts deep Langague models for effective retrieval.

ColBERT introduces a late interaction architecture that independently encodes the query and the document with BERT to attain fine-grained similarity from the intreaction between query and document

If you want to know and understand the detailed information about ColBERT, Read the paper which titled ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT. Khattab et al., arXiv 2020.

Reference

• Paper
  • arXiv Version: ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT. Khattab et al., arXiv 2020
• How to use html for alert
  • how to use icon
• How to use MathJax
  • MathJax basic tutorial and quick reference in StackExchange