Discourse Graph Guided Document Translation with Large Language Models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917071229550592 |
|---|---|
| author | Pham, Viet-Thanh Wang, Minghan Liao, Hao-Han Vu, Thuy-Trang |
| author_facet | Pham, Viet-Thanh Wang, Minghan Liao, Hao-Han Vu, Thuy-Trang |
| contents | Adapting large language models to full document translation remains challenging due to the difficulty of capturing long-range dependencies and preserving discourse coherence throughout extended texts. While recent agentic machine translation systems mitigate context window constraints through multi-agent orchestration and persistent memory, they require substantial computational resources and are sensitive to memory retrieval strategies. We introduce TransGraph, a discourse-guided framework that explicitly models inter-chunk relationships through structured discourse graphs and selectively conditions each translation segment on relevant graph neighbourhoods rather than relying on sequential or exhaustive context. Across three document-level MT benchmarks spanning six languages and diverse domains, TransGraph consistently surpasses strong baselines in translation quality and terminology consistency while incurring significantly lower token overhead. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_07230 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Discourse Graph Guided Document Translation with Large Language Models Pham, Viet-Thanh Wang, Minghan Liao, Hao-Han Vu, Thuy-Trang Computation and Language Artificial Intelligence Adapting large language models to full document translation remains challenging due to the difficulty of capturing long-range dependencies and preserving discourse coherence throughout extended texts. While recent agentic machine translation systems mitigate context window constraints through multi-agent orchestration and persistent memory, they require substantial computational resources and are sensitive to memory retrieval strategies. We introduce TransGraph, a discourse-guided framework that explicitly models inter-chunk relationships through structured discourse graphs and selectively conditions each translation segment on relevant graph neighbourhoods rather than relying on sequential or exhaustive context. Across three document-level MT benchmarks spanning six languages and diverse domains, TransGraph consistently surpasses strong baselines in translation quality and terminology consistency while incurring significantly lower token overhead. |
| title | Discourse Graph Guided Document Translation with Large Language Models |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2511.07230 |