Learn about an innovative approach to dense passage retrieval in this 15-minute conference talk from SIGIR 2024. Explore how pre-training with bag-of-word prediction can enhance retrieval performance by eliminating the need for a decoder component. Discover the technical implementation details and potential benefits of this streamlined approach presented at the Association for Computing Machinery's Special Interest Group on Information Retrieval conference.
Drop Your Decoder: Pre-training with Bag-of-Word Prediction for Dense Passage Retrieval
Association for Computing Machinery (ACM) via YouTube
Overview
Syllabus
SIGIR 2024 W1.1 [fp] Drop your Decoder: Pre-training with Bag-of-Word Prediction
Taught by
Association for Computing Machinery (ACM)
Related Courses
-
Negative Sampling Techniques for Dense Passage Retrieval in a Multilingual Setting - Lecture 4
-
M2-RAAP: A Multi-Modal Recipe for Advancing Adaptation-based Pre-training for Video-text Retrieval
-
Scaling Laws for Dense Retrieval
-
Generative Retrieval as Multi-Vector Dense Retrieval
-
Leveraging LLMs for Unsupervised Dense Retriever Ranking - Dense Retrieval 1
-
Dimension Importance Estimation for Dense Information Retrieval - Tutorial 2.1
