Annif at the GermEval-2025 LLMs4Subjects Task: Traditional XMTC Augmented by Efficient LLMs

TL;DR

This paper describes an enhanced Annif system that combines traditional extreme multi-label classification with efficient large language models to improve subject prediction accuracy and efficiency in bibliographic record indexing.

Contribution

The paper introduces a novel hybrid approach that integrates small, efficient LLMs with traditional methods, achieving top performance in GermEval-2025's subject classification task.

Findings

Ranked 1st in overall quantitative evaluation

Achieved top qualitative evaluation results

Demonstrated improved efficiency and accuracy

Abstract

This paper presents the Annif system in the LLMs4Subjects shared task (Subtask 2) at GermEval-2025. The task required creating subject predictions for bibliographic records using large language models, with a special focus on computational efficiency. Our system, based on the Annif automated subject indexing toolkit, refines our previous system from the first LLMs4Subjects shared task, which produced excellent results. We further improved the system by using many small and efficient language models for translation and synthetic data generation and by using LLMs for ranking candidate subjects. Our system ranked 1st in the overall quantitative evaluation of and 1st in the qualitative evaluation of Subtask 2.