Metraipolis › archive › dc_7ed2d6d27cb3
CNM-BERT: A Drop-In Structural Embedding for Chinese Characters via Ideographic Description Sequences — arXiv:2608.05167
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.05167
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-25T21:27:03.910Z
- SHA-256 of the archived text
e848d21d6e85d7e827e6044e5e4a1762074e9cf85219499428f1bb763039b855- Stable URL of this page
https://metraipolis.com/documento/dc_7ed2d6d27cb3
How to cite
CNM-BERT: A Drop-In Structural Embedding for Chinese Characters via Ideographic Description Sequences — arXiv:2608.05167 — arXiv (en), https://arxiv.org/abs/2608.05167 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7ed2d6d27cb3 (Metraipolis archive dc_7ed2d6d27cb3, sha256 e848d21d6e85d7e827e6044e5e4a1762074e9cf85219499428f1bb763039b855)
BibTeX
@misc{dc7ed2d6d27cb3,
title = {CNM-BERT: A Drop-In Structural Embedding for Chinese Characters via Ideographic Description Sequences — arXiv:2608.05167},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.05167},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7ed2d6d27cb3; sha256 e848d21d6e85d7e827e6044e5e4a1762074e9cf85219499428f1bb763039b855},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05167. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Token-based encoders like BERT treat Chinese characters as atomic identifiers, ignoring their recursive orthographic structure. Consequently, models rely on contextual co-occurrence, degrading performance on rare and out-of-vocabulary (OOV) characters. We propose the Compositional Network Model (CNM), a lightweight augmentation that injects discrete compositional structure into Transformer encoders. CNM parses Ideographic Description Sequences (IDS) into trees, encodes them via a recursive Tree-MLP, and fuses the structural embeddings into BERT without modifying the backbone. Evaluated on the Wu et al. (2025) structural-probing benchmark, CNM-BERT outperforms the strongest baseline (ChineseBERT) on long-tail and OOV characters by +9.8 Structure accuracy and +7.7 Radical F1. Furthermore, CNM-BERT achieves consistent gains across CLUE, MRC, and NER tasks at both base and large scales, demonstrating that explicit structural injection delivers both robust OOV understanding and tangible downstream value. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-07 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.05167 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-25T21:24:38.177Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).