Logits-Constrained Framework with RoBERTa for Ancient Chinese NER

2025/05/05 by Wenjie Hua, Hua, Wenjie, Shenghan Xu +1 · 1 citation
Computer Science · #68T50 #Computation and Language (cs.CL) #FOS: Computer and information sciences #I.2.7 #I.5.1 #I.5.4 #Natural Language Processing Techniques

paper · pdf · doi:10.48550/arxiv.2505.02983

openalex publication_date 2025/05/05 · openalex created_date 2025/10/16 · openalex updated_date 2026/07/28

Abstract

This paper presents a Logits-Constrained (LC) framework for Ancient Chinese Named Entity Recognition (NER), evaluated on the EvaHan 2025 benchmark. Our two-stage model integrates GujiRoBERTa for contextual encoding and a differentiable decoding mechanism to enforce valid BMES label transitions. Experiments demonstrate that LC improves performance over traditional CRF and BiLSTM-based approaches, especially in high-label or large-data settings. We also propose a model selection criterion balancing label complexity and dataset size, providing practical guidance for real-world Ancient Chinese NLP tasks.

Citations

• Named Entity Recognition with Small Strongly Labeled and Large Weakly Labeled Data
• Masked Conditional Random Fields for Sequence Labeling
• RoBERTa: A Robustly Optimized BERT Pretraining Approach
• Bidirectional LSTM-CRF Models for Sequence Tagging
• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Cited by

• BIRD: Bronze Inscription Restoration and Dating

Related

• Using Chinese Glyphs for Named Entity Recognition
• A More Efficient Chinese Named Entity Recognition base on BERT and Syntactic Analysis
• CHisIEC: An Information Extraction Corpus for Ancient Chinese History
• CLUENER2020: Fine-grained Named Entity Recognition Dataset and Benchmark for Chinese
• Simplify the Usage of Lexicon in Chinese NER