Publication details
The Art of Hierarchical Competing Patterns: Gaussian Process Optimization of Hyphenation
| Authors | |
|---|---|
| Year of publication | 2026 |
| Type | Paper in proceedings |
| Conference | Findings of the Association for Computational Linguistics: EMNLP 2026 |
| MU Faculty or unit | |
| Citation | |
| web | |
| Doi | https://doi.org/10.18653/2026.findings-emnlp.2346 |
| Keywords | Gaussian process; optimization; hyphenation; competing patterns; pattern generation; segmentation |
| Description | Hyphenation patterns remain a compact and widely deployed solution for word breaking in typesetting systems, text processors, and web rendering engines, but their generation still depends on manually tuned patgen program parameter profiles. We formulate patgen profile selection as a black-box hyperparameter optimization problem and evaluate Gaussian-process Bayesian optimization for this task. The search objective combines a precision-oriented $F_{1/7}$ score with an explicit trie size-accuracy tradeoff using a normalized trie-size penalty. We evaluate the method on 17 hyphenated word-list datasets covering 14 languages and multiple scripts. Against two strong hand-tuned profiles regenerated from the same 8/10 training split and evaluated on the same 1/10 held-out test split, the GP-optimized profiles improve $F_{1/7}$ and reduce trie size on all 17 datasets. The median optimized/baseline trie ratio is 0.348. A dataset-level sign test gives $p=7.6\cdot10^{-6}$; a separate budget-matched comparison on five representative datasets shows that model-based search improves over the best hand-tuned profile under the fixed comparison objective. The results show that model-based optimization can make pattern generation more reproducible and less dependent on expert trial-and-error while keeping the accuracy--compactness trade-off explicit. |
| Related projects: |