Does generative AI supersede supervised XMLC? A Benchmark Study on Automated Subject Indexing with German Scientific Literature
Authors: Maximilian Kähler, Katja Konermann, Lisa Kluge, Markus Schumacher
Organizations: Deutsche Nationalbibliothek Leipzig/ Frankfurt am Main, Germany
Abstract
With a large controlled vocabulary as the label set, the task of automated subject indexing in a library can be understood as a multi-label classification task. If the set of subject terms is large, the problem fits the Extreme Multi-Label Classification (XMLC) objective. In this study, we apply a selection of specialised supervised XMLC methods to the test case of subject indexing contemporary German scientific literature, collected at the German National Library (DNB). We contrast these results by including a classical lexical matching baseline and three of our own recently developed LLM-based methods into the benchmark. Algorithms are evaluated and compared in several metrics. This includes binary relevance comparisons with previously indexed material, as well as graded relevance ratings by professional subject librarians. A challenge for all methods is to reliably make suggestions from the long tail of the subject vocabulary. We find that supervised XMLC algorithms relying on transformer-based dense features give best results in terms of overall binary relevance metrics. However, focusing on graded relevance and performance in the long tail of our subject vocabulary, the LLM-based generative methods give better results, making them a promising alternative for future productive use.
This paper presents a modular AI agentic skill pipeline for automating subject indexing with Library of Congress Subject Headings (LCSH). Subject indexing - the process of analyzing a work's aboutness, selecting controlled vocabulary terms, and encoding them as MARC21 subject access fields - is one of the most time-consuming components of library cataloging. The system decomposes this process into four discrete, sequentially executed agent skills: conceptual analysis, quantitative filtering, authority validation, and MARC field synthesis. Each skill encodes domain knowledge drawn directly from Library of Congress Subject Headings Manual (SHM) instruction sheets and subject analysis theory. The pipeline was evaluated against a corpus of ten titles whose existing subject headings were captured from the Harvard Library bibliographic dataset (a snapshot of their Alma ILS). Results demonstrate strong conceptual alignment with professional subject indexing practice, with notable differences in specificity, subdivision practice, and the agent's adherence to the 2026 LC policy discontinuing form subdivisions in favor of LCGFT 655 fields.
Automated subject cataloging assigns controlledvocabulary headings to bibliographic records, but LCSH has no standard public benchmark. We introduce LCSHBench: 22,346 books in 15 languages from the openly licensed Harvard, Columbia, and Princeton catalogs. Records enter only when at least two independent cataloging agencies assigned LCSH; we release per-catalog provenance plus union and unanimous answer views. A concordance study of 465,187 works cataloged by all three libraries shows why this design matters: libraries usually agree on the underlying topic (93.3% share a concept-level heading) but often differ in exact expression (39.4% have identical heading sets). LCSHBench therefore scores both exact and concept matches, with set and rank metrics broken down by language and heading type, across open-vocabulary generation and full-vocabulary retrieval. As a first demonstration, a low-rank fine-tune of a 300M on-device embedder improves cross-lingual retrieval and beats a 3,072-dimensional hosted embedder on development exact recall@200 (0.659 vs 0.623). The language panel shows the gain is not uniform, and held-out-test and end-to-end confirmation remain future work.
Libraries and archives manage large collections with limited staff and computing budgets, yet common benchmarks do not systematically test their bibliographic work. They need to know which methods work for their tasks and what those methods require to run. SHELF, the Synthetic Harness for Evaluating LLM Fitness, addresses this gap. It is a Python system that turns labelled taxonomies, writing specifications, and a generation budget into controlled benchmark data and evaluation tasks. This first release contains 62,899 model-written documents based on Library of Congress vocabularies, with tasks for classification, clustering, retrieval, pair classification, and instruction retrieval. We compare TF, TF-IDF, BM25, popular encoders, and, on subject classification only, zero-shot decoders; each method appears only on tasks that support it. Subject classification reaches 0.8887, while genre-form classification reaches only 0.2605, and several pair and clustering tasks remain near chance. Sparse methods remain competitive on classification, while TF-IDF is the fastest measured arm in the subject timing experiment. SHELF also varies bibliographic facets independently and can generate new, verifiably unseen documents after a model's training cutoff. Comparisons with LCSHBench and Project Gutenberg show that model rankings transfer more reliably than absolute scores, but SHELF scores do not estimate accuracy on production catalogue data. We release all source code and data under permissive licenses on GitHub and Hugging Face.