When Less Is More: Logits-Constrained Framework with RoBERTa for Ancient Chinese NER
Wenjie Hua, Shenghan Xu · 2025
This report presents our team's work on ancient Chinese Named Entity Recognition (NER) for EvaHan 2025 1 .We propose a two-stage framework combining GujiRoBERTa with a Logits-Constrained (LC) mechanism.The first stage generates contextual embeddings using Gu-jiRoBERTa, followed by dynamically masked decoding to enforce valid BMES transitions.Experiments on EvaHan 2025 datasets demonstrate the framework's effectiveness.Key findings include the LC framework's superiority over CRFs in high-label scenarios and the detrimental effect of BiLSTM modules.We also establish empirical model selection guidelines based on label complexity and dataset size.