Saved in:
Bibliographic Details
Main Authors: Iturra-Bocaz, Gabriel, Vo, Danny, Galuscakova, Petra
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.23191
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913917438001152
author Iturra-Bocaz, Gabriel
Vo, Danny
Galuscakova, Petra
author_facet Iturra-Bocaz, Gabriel
Vo, Danny
Galuscakova, Petra
contents This paper investigates the impact of shallow versus deep relevance judgments on the performance of BERT-based reranking models in neural Information Retrieval. Shallow-judged datasets, characterized by numerous queries each with few relevance judgments, and deep-judged datasets, involving fewer queries with extensive relevance judgments, are compared. The research assesses how these datasets affect the performance of BERT-based reranking models trained on them. The experiments are run on the MS MARCO and LongEval collections. Results indicate that shallow-judged datasets generally enhance generalization and effectiveness of reranking models due to a broader range of available contexts. The disadvantage of the deep-judged datasets might be mitigated by a larger number of negative training examples.
format Preprint
id arxiv_https___arxiv_org_abs_2506_23191
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Impact of Shallow vs. Deep Relevance Judgments on BERT-based Reranking Models
Iturra-Bocaz, Gabriel
Vo, Danny
Galuscakova, Petra
Information Retrieval
This paper investigates the impact of shallow versus deep relevance judgments on the performance of BERT-based reranking models in neural Information Retrieval. Shallow-judged datasets, characterized by numerous queries each with few relevance judgments, and deep-judged datasets, involving fewer queries with extensive relevance judgments, are compared. The research assesses how these datasets affect the performance of BERT-based reranking models trained on them. The experiments are run on the MS MARCO and LongEval collections. Results indicate that shallow-judged datasets generally enhance generalization and effectiveness of reranking models due to a broader range of available contexts. The disadvantage of the deep-judged datasets might be mitigated by a larger number of negative training examples.
title Impact of Shallow vs. Deep Relevance Judgments on BERT-based Reranking Models
topic Information Retrieval
url https://arxiv.org/abs/2506.23191