Multimodal learned sparse retrieval for image suggestion

T Nguyen, M Hendriksen, A Yates - arXiv preprint arXiv:2402.07736, 2024 - arxiv.org
Learned Sparse Retrieval (LSR) is a group of neural methods designed to encode queries
and documents into sparse lexical vectors. These vectors can be efficiently indexed and …

Assessing Brittleness of Image-Text Retrieval Benchmarks from Vision-Language Models Perspective

M Hendriksen, S Zhang, R Reinanda, M Yahya… - arXiv preprint arXiv …, 2024 - arxiv.org
We examine the brittleness of the image-text retrieval (ITR) evaluation pipeline with a focus
on concept granularity. We start by analyzing two common benchmarks, MS-COCO and …

DyVo: Dynamic Vocabularies for Learned Sparse Retrieval with Entities

T Nguyen, S Chatterjee, S MacAvaney, I Mackie… - arXiv preprint arXiv …, 2024 - arxiv.org
Learned Sparse Retrieval (LSR) models use vocabularies from pre-trained transformers,
which often split entities into nonsensical fragments. Splitting entities can reduce retrieval …

Neural Lexical Search with Learned Sparse Retrieval

A Yates, C Lassance, S MacAvaney… - Proceedings of the 2024 …, 2024 - dl.acm.org
Learned Sparse Retrieval (LSR) techniques use neural machinery to represent queries and
documents as learned bags of words. In contrast with other neural retrieval techniques, such …

Retrieval Evaluation for Long-Form and Knowledge-Intensive Image–Text Article Composition

JH Yang, C Lassance, R Rezende… - Proceedings of the …, 2024 - aclanthology.org
This paper examines the integration of images into Wikipedia articles by evaluating image–
text retrieval tasks in multimedia content creation, focusing on developing retrieval …