Duluth at SemEval-2025 Task 7: TF-IDF with Optimized Vector Dimensions for Multilingual Fact-Checked Claim Retrieval

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Syed, Shujauddin, Pedersen, Ted
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909615181004800
author Syed, Shujauddin
Pedersen, Ted
author_facet Syed, Shujauddin
Pedersen, Ted
contents This paper presents the Duluth approach to the SemEval-2025 Task 7 on Multilingual and Crosslingual Fact-Checked Claim Retrieval. We implemented a TF-IDF-based retrieval system with experimentation on vector dimensions and tokenization strategies. Our best-performing configuration used word-level tokenization with a vocabulary size of 15,000 features, achieving an average success@10 score of 0.78 on the development set and 0.69 on the test set across ten languages. Our system showed stronger performance on higher-resource languages but still lagged significantly behind the top-ranked system, which achieved 0.96 average success@10. Our findings suggest that though advanced neural architectures are increasingly dominant in multilingual retrieval tasks, properly optimized traditional methods like TF-IDF remain competitive baselines, especially in limited compute resource scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2505_12616
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Duluth at SemEval-2025 Task 7: TF-IDF with Optimized Vector Dimensions for Multilingual Fact-Checked Claim Retrieval
Syed, Shujauddin
Pedersen, Ted
Computation and Language
68T50
This paper presents the Duluth approach to the SemEval-2025 Task 7 on Multilingual and Crosslingual Fact-Checked Claim Retrieval. We implemented a TF-IDF-based retrieval system with experimentation on vector dimensions and tokenization strategies. Our best-performing configuration used word-level tokenization with a vocabulary size of 15,000 features, achieving an average success@10 score of 0.78 on the development set and 0.69 on the test set across ten languages. Our system showed stronger performance on higher-resource languages but still lagged significantly behind the top-ranked system, which achieved 0.96 average success@10. Our findings suggest that though advanced neural architectures are increasingly dominant in multilingual retrieval tasks, properly optimized traditional methods like TF-IDF remain competitive baselines, especially in limited compute resource scenarios.
title Duluth at SemEval-2025 Task 7: TF-IDF with Optimized Vector Dimensions for Multilingual Fact-Checked Claim Retrieval
topic Computation and Language
68T50
url https://arxiv.org/abs/2505.12616