Logits-Constrained Framework with RoBERTa for Ancient Chinese NER

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hua, Wenjie, Xu, Shenghan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915275188731904
author Hua, Wenjie
Xu, Shenghan
author_facet Hua, Wenjie
Xu, Shenghan
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2505_02983
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Logits-Constrained Framework with RoBERTa for Ancient Chinese NER
Hua, Wenjie
Xu, Shenghan
Computation and Language
68T50
I.2.7; I.5.1; I.5.4
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.
title Logits-Constrained Framework with RoBERTa for Ancient Chinese NER
topic Computation and Language
68T50
I.2.7; I.5.1; I.5.4
url https://arxiv.org/abs/2505.02983