DNB-AI-Project at SemEval-2025 Task 5: An LLM-Ensemble Approach for Automated Subject Indexing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kluge, Lisa, Kähler, Maximilian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918124720226304
author Kluge, Lisa
Kähler, Maximilian
author_facet Kluge, Lisa
Kähler, Maximilian
contents This paper presents our system developed for the SemEval-2025 Task 5: LLMs4Subjects: LLM-based Automated Subject Tagging for a National Technical Library's Open-Access Catalog. Our system relies on prompting a selection of LLMs with varying examples of intellectually annotated records and asking the LLMs to similarly suggest keywords for new records. This few-shot prompting technique is combined with a series of post-processing steps that map the generated keywords to the target vocabulary, aggregate the resulting subject terms to an ensemble vote and, finally, rank them as to their relevance to the record. Our system is fourth in the quantitative ranking in the all-subjects track, but achieves the best result in the qualitative ranking conducted by subject indexing experts.
format Preprint
id arxiv_https___arxiv_org_abs_2504_21589
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DNB-AI-Project at SemEval-2025 Task 5: An LLM-Ensemble Approach for Automated Subject Indexing
Kluge, Lisa
Kähler, Maximilian
Computation and Language
Artificial Intelligence
Digital Libraries
I.2.7
This paper presents our system developed for the SemEval-2025 Task 5: LLMs4Subjects: LLM-based Automated Subject Tagging for a National Technical Library's Open-Access Catalog. Our system relies on prompting a selection of LLMs with varying examples of intellectually annotated records and asking the LLMs to similarly suggest keywords for new records. This few-shot prompting technique is combined with a series of post-processing steps that map the generated keywords to the target vocabulary, aggregate the resulting subject terms to an ensemble vote and, finally, rank them as to their relevance to the record. Our system is fourth in the quantitative ranking in the all-subjects track, but achieves the best result in the qualitative ranking conducted by subject indexing experts.
title DNB-AI-Project at SemEval-2025 Task 5: An LLM-Ensemble Approach for Automated Subject Indexing
topic Computation and Language
Artificial Intelligence
Digital Libraries
I.2.7
url https://arxiv.org/abs/2504.21589