Zobrazeno 1 - 7
of 7
pro vyhledávání: '"Zhu Wenhao"'
The paradigm of pre-training followed by fine-tuning on downstream tasks has become the mainstream method in natural language processing tasks. Although pre-trained models have the advantage of generalization, their performance may still vary signifi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4a56337134d861891d92047329cec678
http://arxiv.org/abs/2306.12280
http://arxiv.org/abs/2306.12280
Autor:
Zhu, Wenhao, Liu, Hongyi, Dong, Qingxiu, Xu, Jingjing, Huang, Shujian, Kong, Lingpeng, Chen, Jiajun, Li, Lei
Large language models (LLMs) have demonstrated remarkable potential in handling multilingual machine translation (MMT). In this paper, we systematically investigate the advantages and challenges of LLMs for MMT by answering two questions: 1) How well
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ea05bb5c6bb80dbeaaf35353829a239b
http://arxiv.org/abs/2304.04675
http://arxiv.org/abs/2304.04675
Autor:
Zhu, Wenhao, Zhao, Qianfeng, Lv, Yunzhe, Huang, Shujian, Zhao, Siheng, Liu, Sizhe, Chen, Jiajun
Augmenting the base neural model with a token-level symbolic datastore is a novel generation paradigm and has achieved promising results in machine translation (MT). In this paper, we introduce a unified framework kNN-BOX, which enables quick develop
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b8b140f7bc495d74de5b26a2f988a69a
http://arxiv.org/abs/2302.13574
http://arxiv.org/abs/2302.13574
Neural machine translation has achieved promising results on many translation tasks. However, previous studies have shown that neural models induce a non-smooth representation space, which harms its generalization results. Recently, kNN-MT has provid
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2b629d8aecc06c35f49cf044592158e5
Multilingual neural machine translation (MNMT) aims to build a unified model for many language directions. Existing monolithic models for MNMT encounter two challenges: parameter interference among languages and inefficient inference for large models
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9fdc1214faa4ab05a4b28e8b79315f3c
kNN-MT presents a new paradigm for domain adaptation by building an external datastore, which usually saves all target language token occurrences in the parallel corpus. As a result, the constructed datastore is usually large and possibly redundant.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::25b76c4dba607f524d13ec69d1143adc
Autor:
Zhu, Wenhao, Huang, Shujian, Pu, Tong, Huang, Pingxuan, Zhang, Xu, Yu, Jian, Chen, Wei, Wang, Yanfeng, Chen, Jiajun
Previous research for adapting a general neural machine translation (NMT) model into a specific domain usually neglects the diversity in translation within the same domain, which is a core problem for domain adaptation in real-world scenarios. One re
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4f97080c5faf003fbb112fdd52f93097
http://arxiv.org/abs/2012.15717
http://arxiv.org/abs/2012.15717