Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.01589 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914132765179904 |
|---|---|
| author | Hua, Wenjie Nguyen, Hoang H. Ge, Gangyan |
| author_facet | Hua, Wenjie Nguyen, Hoang H. Ge, Gangyan |
| contents | Bronze inscriptions from early China are fragmentary and difficult to date. We introduce BIRD(Bronze Inscription Restoration and Dating), a fully encoded dataset grounded in standard scholarly transcriptions and chronological labels. We further propose an allograph-aware masked language modeling framework that integrates domain- and task-adaptive pretraining with a Glyph Net (GN), which links graphemes and allographs. Experiments show that GN improves restoration, while glyph-biased sampling yields gains in dating. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_01589 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | BIRD: Bronze Inscription Restoration and Dating Hua, Wenjie Nguyen, Hoang H. Ge, Gangyan Computation and Language I.2.7 Bronze inscriptions from early China are fragmentary and difficult to date. We introduce BIRD(Bronze Inscription Restoration and Dating), a fully encoded dataset grounded in standard scholarly transcriptions and chronological labels. We further propose an allograph-aware masked language modeling framework that integrates domain- and task-adaptive pretraining with a Glyph Net (GN), which links graphemes and allographs. Experiments show that GN improves restoration, while glyph-biased sampling yields gains in dating. |
| title | BIRD: Bronze Inscription Restoration and Dating |
| topic | Computation and Language I.2.7 |
| url | https://arxiv.org/abs/2511.01589 |