Revisiting the Markov Property for Machine Translation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913222429245440 |
|---|---|
| author | Du, Cunxiao Zhou, Hao Tu, Zhaopeng Jiang, Jing |
| author_facet | Du, Cunxiao Zhou, Hao Tu, Zhaopeng Jiang, Jing |
| contents | In this paper, we re-examine the Markov property in the context of neural machine translation. We design a Markov Autoregressive Transformer~(MAT) and undertake a comprehensive assessment of its performance across four WMT benchmarks. Our findings indicate that MAT with an order larger than 4 can generate translations with quality on par with that of conventional autoregressive transformers. In addition, counter-intuitively, we also find that the advantages of utilizing a higher-order MAT do not specifically contribute to the translation of longer sentences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_02084 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Revisiting the Markov Property for Machine Translation Du, Cunxiao Zhou, Hao Tu, Zhaopeng Jiang, Jing Computation and Language In this paper, we re-examine the Markov property in the context of neural machine translation. We design a Markov Autoregressive Transformer~(MAT) and undertake a comprehensive assessment of its performance across four WMT benchmarks. Our findings indicate that MAT with an order larger than 4 can generate translations with quality on par with that of conventional autoregressive transformers. In addition, counter-intuitively, we also find that the advantages of utilizing a higher-order MAT do not specifically contribute to the translation of longer sentences. |
| title | Revisiting the Markov Property for Machine Translation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2402.02084 |