What Have We Achieved on Non-autoregressive Translation?
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913357945110528 |
|---|---|
| author | Li, Yafu Zhang, Huajian Yan, Jianhao Yin, Yongjing Zhang, Yue |
| author_facet | Li, Yafu Zhang, Huajian Yan, Jianhao Yin, Yongjing Zhang, Yue |
| contents | Recent advances have made non-autoregressive (NAT) translation comparable to autoregressive methods (AT). However, their evaluation using BLEU has been shown to weakly correlate with human annotations. Limited research compares non-autoregressive translation and autoregressive translation comprehensively, leaving uncertainty about the true proximity of NAT to AT. To address this gap, we systematically evaluate four representative NAT methods across various dimensions, including human evaluation. Our empirical results demonstrate that despite narrowing the performance gap, state-of-the-art NAT still underperforms AT under more reliable evaluation metrics. Furthermore, we discover that explicitly modeling dependencies is crucial for generating natural language and generalizing to out-of-distribution sequences. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_12788 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | What Have We Achieved on Non-autoregressive Translation? Li, Yafu Zhang, Huajian Yan, Jianhao Yin, Yongjing Zhang, Yue Computation and Language Recent advances have made non-autoregressive (NAT) translation comparable to autoregressive methods (AT). However, their evaluation using BLEU has been shown to weakly correlate with human annotations. Limited research compares non-autoregressive translation and autoregressive translation comprehensively, leaving uncertainty about the true proximity of NAT to AT. To address this gap, we systematically evaluate four representative NAT methods across various dimensions, including human evaluation. Our empirical results demonstrate that despite narrowing the performance gap, state-of-the-art NAT still underperforms AT under more reliable evaluation metrics. Furthermore, we discover that explicitly modeling dependencies is crucial for generating natural language and generalizing to out-of-distribution sequences. |
| title | What Have We Achieved on Non-autoregressive Translation? |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2405.12788 |