On the Reproducibility of Learned Sparse Retrieval Adaptations for Long Documents

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lionis, Emmanouil Georgios, Ju, Jia-Huei
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909571176464384
author Lionis, Emmanouil Georgios
Ju, Jia-Huei
author_facet Lionis, Emmanouil Georgios
Ju, Jia-Huei
contents Document retrieval is one of the most challenging tasks in Information Retrieval. It requires handling longer contexts, often resulting in higher query latency and increased computational overhead. Recently, Learned Sparse Retrieval (LSR) has emerged as a promising approach to address these challenges. Some have proposed adapting the LSR approach to longer documents by aggregating segmented document using different post-hoc methods, including n-grams and proximity scores, adjusting representations, and learning to ensemble all signals. In this study, we aim to reproduce and examine the mechanisms of adapting LSR for long documents. Our reproducibility experiments confirmed the importance of specific segments, with the first segment consistently dominating document retrieval performance. Furthermore, We re-evaluate recently proposed methods -- ExactSDM and SoftSDM -- across varying document lengths, from short (up to 2 segments) to longer (3+ segments). We also designed multiple analyses to probe the reproduced methods and shed light on the impact of global information on adapting LSR to longer contexts. The complete code and implementation for this project is available at: https://github.com/lionisakis/Reproducibilitiy-lsr-long.
format Preprint
id arxiv_https___arxiv_org_abs_2503_23824
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the Reproducibility of Learned Sparse Retrieval Adaptations for Long Documents
Lionis, Emmanouil Georgios
Ju, Jia-Huei
Information Retrieval
Document retrieval is one of the most challenging tasks in Information Retrieval. It requires handling longer contexts, often resulting in higher query latency and increased computational overhead. Recently, Learned Sparse Retrieval (LSR) has emerged as a promising approach to address these challenges. Some have proposed adapting the LSR approach to longer documents by aggregating segmented document using different post-hoc methods, including n-grams and proximity scores, adjusting representations, and learning to ensemble all signals. In this study, we aim to reproduce and examine the mechanisms of adapting LSR for long documents. Our reproducibility experiments confirmed the importance of specific segments, with the first segment consistently dominating document retrieval performance. Furthermore, We re-evaluate recently proposed methods -- ExactSDM and SoftSDM -- across varying document lengths, from short (up to 2 segments) to longer (3+ segments). We also designed multiple analyses to probe the reproduced methods and shed light on the impact of global information on adapting LSR to longer contexts. The complete code and implementation for this project is available at: https://github.com/lionisakis/Reproducibilitiy-lsr-long.
title On the Reproducibility of Learned Sparse Retrieval Adaptations for Long Documents
topic Information Retrieval
url https://arxiv.org/abs/2503.23824