arXiv：cs.AI· Miguel Arana-Catania, Gillian Pink, Glenn Roe·· 5 小时前
用机器学习为伏尔泰全集自动编制主题索引：Mistral 4-bit 量化模型 F1 最高 0.67
Automatic Thematic Indexing of Large Literary Corpora: A Machine Learning Approach to Voltaire's Complete Works
AI 导读
研究将伏尔泰《风俗论》与《百科全书问题》的主题索引编制建模为多标签分类任务，比较了带分类头的编码器模型与经 LoRA 微调的生成式 LLM，覆盖约 3B 至 120B 参数规模。表现最好的 Mistral 家族模型在 4-bit 量化配置下 F1 最高达 0.67，作者认为受专业索引主观性影响，该数值属于下界。研究还评估了跨语料泛化能力，并定性分析了模型在文学与修辞特征上的失效表现。
正文
Abstract:Thematic indexing -- the practice of assigning structured conceptual labels to sections of text -- is essential to scholarly access in large-scale literary and historical editions, yet it remains a largely manual, labour-intensive process. This paper explores the application of machine learning to automatic thematic indexing, using two substantial sub-corpora of the Complete Works of Voltaire as a test case: the Essai sur les mœurs et l'esprit des nations and the Questions sur l'Encyclopédie. The task is framed as a multi-label classification problem, in which a model must assign the set of index entries that a professional indexer would apply to a given page of text. We compare a range of approaches -- from encoder-based models with classification heads to generative large language models (LLMs) fine-tuned via Low-Rank Adaptation (LoRA) -- spanning model sizes from approximately 3 to 120 billion parameters. Our best-performing model, from the Mistral family in a 4-bit quantised configuration, achieves F1 scores of up to 0.67; we argue that these figures represent lower bounds, given the inherent subjectivity of professional indexing and the frequency with which model predictions prove semantically valid despite diverging from the print index. We further evaluate cross-corpus generalisation and conduct a detailed qualitative analysis of model behaviour on literary and rhetorical features of the source texts that prove particularly resistant to automated treatment. Our findings have implications for the broader challenge of providing structured thematic access to large-scale literary and historical corpora.
| Comments: | 26 pages, 3 figures, 3 tables |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Digital Libraries (cs.DL); Information Retrieval (cs.IR); Machine Learning (cs.LG) |
| Cite as: | arXiv:2607.09316 [cs.CL] |
| (or arXiv:2607.09316v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.09316 arXiv-issued DOI via DataCite |
Submission history
From: Miguel Arana-Catania [view email]
[v1]
Fri, 10 Jul 2026 11:54:02 UTC (576 KB)
[v2]
Tue, 13 Oct 2026 08:58:58 UTC (582 KB)
来源：arXiv：cs.AI · arxiv.org