Metraipolis › archive › dc_2d32fd9817ec
Scaling Unsupervised Word Alignment to Documents via Structural Constraints — arXiv:2608.21023
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.21023
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T10:18:28.445Z
- SHA-256 of the archived text
ed64a804508aba467e0744c25e91e227464da1a69c7872728dd60b3da590a725- Stable URL of this page
https://metraipolis.com/documento/dc_2d32fd9817ec
How to cite
Scaling Unsupervised Word Alignment to Documents via Structural Constraints — arXiv:2608.21023 — arXiv (en), https://arxiv.org/abs/2608.21023 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2d32fd9817ec (Metraipolis archive dc_2d32fd9817ec, sha256 ed64a804508aba467e0744c25e91e227464da1a69c7872728dd60b3da590a725)
BibTeX
@misc{dc2d32fd9817ec,
title = {Scaling Unsupervised Word Alignment to Documents via Structural Constraints — arXiv:2608.21023},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.21023},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2d32fd9817ec; sha256 ed64a804508aba467e0744c25e91e227464da1a69c7872728dd60b3da590a725},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.21023. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Word alignment has traditionally been studied between sentences, but many cross-lingual tasks increasingly require correspondences across full documents. While recent multilingual embedding models can encode long inputs, we show that applying algorithms designed for sentences directly to documents leads to performance degradation. To address this, we introduce CTFAlign, a lightweight, training-free approach for document-level word alignment. CTFAlign applies a coarse-to-fine refinement strategy that restricts the alignment search space to semantically similar regions. Additionally, we introduce MDPAlign, a simpler alternative that constrains alignments by position with a main diagonal prior. Both approaches operate directly on full documents without relying on sentence segmentation or sentence alignment. We evaluate these methods across six language pairs varying in typological distance, resourcedness, and document length. Averaged over three models, CTFAlign reduces word alignment error rate from 0.412 to 0.326. These gains transfer downstream, leading to improvements in document-level translation coverage evaluation and recognition of semantic differences. We release CTFAlign as a Python package and make the code and data to reproduce our experiments publicly available. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-24 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.21023 — 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 2026-10-18T10:17:37.594Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).