BERT classifier identifies 55,000 letters in Classical Chinese texts

Researchers have developed Lepton, a BERT-based classifier designed to distinguish personal letter titles from prefaces in Classical Chinese collected works. The model was fine-tuned on over 5,000 hand-labeled titles from the late Ming and early Qing dynasties. Lepton has been deployed and used to identify approximately 55,000 letters, contributing to the Ming Letter Platform. AI

Summary written by gemini-2.5-flash-lite from 1 sources. How we write summaries →

IMPACT Enables automated identification of historical correspondence, aiding digital humanities research and archival efforts.

RANK_REASON The cluster describes an academic paper detailing a fine-tuned BERT model for a specific text classification task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• BERT
• Classical Chinese
• Hugging Face
• late Ming and early Qing dynasties
• Lepton
• Ming Letter Platform

• paper
• product