GloCTM: Cross-Lingual Topic Modeling via a Global Context Space
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917207459495936 |
|---|---|
| author | Phat, Nguyen Tien Minh, Ngo Vu Van Ngo, Linh Diep, Nguyen Thi Ngoc Nguyen, Thien Huu |
| author_facet | Phat, Nguyen Tien Minh, Ngo Vu Van Ngo, Linh Diep, Nguyen Thi Ngoc Nguyen, Thien Huu |
| contents | Cross-lingual topic modeling seeks to uncover coherent and semantically aligned topics across languages - a task central to multilingual understanding. Yet most existing models learn topics in disjoint, language-specific spaces and rely on alignment mechanisms (e.g., bilingual dictionaries) that often fail to capture deep cross-lingual semantics, resulting in loosely connected topic spaces. Moreover, these approaches often overlook the rich semantic signals embedded in multilingual pretrained representations, further limiting their ability to capture fine-grained alignment. We introduce GloCTM (Global Context Space for Cross-Lingual Topic Model), a novel framework that enforces cross-lingual topic alignment through a unified semantic space spanning the entire model pipeline. GloCTM constructs enriched input representations by expanding bag-of-words with cross-lingual lexical neighborhoods, and infers topic proportions using both local and global encoders, with their latent representations aligned through internal regularization. At the output level, the global topic-word distribution, defined over the combined vocabulary, structurally synchronizes topic meanings across languages. To further ground topics in deep semantic space, GloCTM incorporates a Centered Kernel Alignment (CKA) loss that aligns the latent topic space with multilingual contextual embeddings. Experiments across multiple benchmarks demonstrate that GloCTM significantly improves topic coherence and cross-lingual alignment, outperforming strong baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_11872 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GloCTM: Cross-Lingual Topic Modeling via a Global Context Space Phat, Nguyen Tien Minh, Ngo Vu Van Ngo, Linh Diep, Nguyen Thi Ngoc Nguyen, Thien Huu Computation and Language Cross-lingual topic modeling seeks to uncover coherent and semantically aligned topics across languages - a task central to multilingual understanding. Yet most existing models learn topics in disjoint, language-specific spaces and rely on alignment mechanisms (e.g., bilingual dictionaries) that often fail to capture deep cross-lingual semantics, resulting in loosely connected topic spaces. Moreover, these approaches often overlook the rich semantic signals embedded in multilingual pretrained representations, further limiting their ability to capture fine-grained alignment. We introduce GloCTM (Global Context Space for Cross-Lingual Topic Model), a novel framework that enforces cross-lingual topic alignment through a unified semantic space spanning the entire model pipeline. GloCTM constructs enriched input representations by expanding bag-of-words with cross-lingual lexical neighborhoods, and infers topic proportions using both local and global encoders, with their latent representations aligned through internal regularization. At the output level, the global topic-word distribution, defined over the combined vocabulary, structurally synchronizes topic meanings across languages. To further ground topics in deep semantic space, GloCTM incorporates a Centered Kernel Alignment (CKA) loss that aligns the latent topic space with multilingual contextual embeddings. Experiments across multiple benchmarks demonstrate that GloCTM significantly improves topic coherence and cross-lingual alignment, outperforming strong baselines. |
| title | GloCTM: Cross-Lingual Topic Modeling via a Global Context Space |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.11872 |