Explore negative sampling techniques for dense passage retrieval in multilingual settings through this 14-minute conference talk presented at SIGIR 2024. Delve into the research conducted by Thilina Rajapakse, Andrew Yates, and Maarten de Rijke as they discuss innovative approaches to improve information retrieval across multiple languages. Gain insights into the challenges and solutions associated with multilingual dense passage retrieval, and learn how negative sampling techniques can enhance the performance of retrieval systems in diverse linguistic contexts.
Negative Sampling Techniques for Dense Passage Retrieval in a Multilingual Setting - Lecture 4
Association for Computing Machinery (ACM) via YouTube
Overview
Syllabus
SIGIR 2024 M2.4 [rr] Negative Sampling Techniques for Dense Passage Retrieval in a Multilingual Set
Taught by
Association for Computing Machinery (ACM)
Related Courses
-
Implementing Multilingual Generative AI Cross-lingual RAGs
-
Generative Retrieval as Multi-Vector Dense Retrieval
-
Multilingual Meta-Distillation Alignment for Semantic Retrieval - Session M2.4
-
Drop Your Decoder: Pre-training with Bag-of-Word Prediction for Dense Passage Retrieval
-
Scaling Laws for Dense Retrieval
-
Steering Large Language Models for Cross-lingual Information Retrieval - Multilingual Retrieval
