Logits-Constrained Framework with RoBERTa for Ancient Chinese NER
This addresses the problem of entity recognition in Ancient Chinese texts, which is an incremental improvement for domain-specific NLP applications.
The paper tackles Named Entity Recognition in Ancient Chinese by proposing a Logits-Constrained framework that integrates GujiRoBERTa with a differentiable decoding mechanism, achieving improved performance over traditional CRF and BiLSTM-based approaches on the EvaHan 2025 benchmark.
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.