Researchers have developed a machine learning approach for automatic thematic indexing of large literary corpora, using the complete works of Voltaire as a test case. The study frames this task as a multi-label classification problem, comparing various models including encoder-based and generative LLMs fine-tuned with LoRA. The best-performing model, a quantized Mistral variant, achieved F1 scores up to 0.67, demonstrating potential for structured thematic access to historical texts. AI
IMPACT This research demonstrates how LLMs can be applied to complex literary analysis, potentially streamlining scholarly access to historical texts.
RANK_REASON Academic paper detailing a machine learning approach to thematic indexing of literary works.
- arXiv
- Essai sur les mœurs et l'esprit des nations
- Hugging Face
- Lora
- Miguel Arana-Catania
- mistral.ai
- Questions sur l'Encyclopédie
- Voltaire
- alphaXiv
- CatalyzeX
- DagsHub
- Gotit.pub
- ScienceCast
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
