Zobrazeno 1 - 7
of 7
pro vyhledávání: '"Huang, Yichong"'
Autor:
Fu, Chengpeng, Feng, Xiaocheng, Huang, Yichong, Huo, Wenshuai, Li, Baohang, Wang, Hui, Qin, Bin, Liu, Ting
Leveraging large language models for machine translation has demonstrated promising results. However, it does require the large language models to possess the capability of handling both the source and target languages in machine translation. When it
Externí odkaz:
http://arxiv.org/abs/2405.02933
Large language models (LLMs) exhibit complementary strengths in various tasks, motivating the research of LLM ensembling. However, existing work focuses on training an extra reward model or fusion model to select or combine all candidate answers, pos
Externí odkaz:
http://arxiv.org/abs/2404.12715
Autor:
Huang, Yichong, Li, Baohang, Feng, Xiaocheng, Fu, Chengpeng, Huo, Wenshuai, Liu, Ting, Qin, Bing
Large Language models (LLMs) have exhibited remarkable abilities in understanding complex texts, offering a promising path towards human-like translation performance. However, this study reveals the misalignment between the translation-specific under
Externí odkaz:
http://arxiv.org/abs/2401.05072
Multilingual neural machine translation has witnessed remarkable progress in recent years. However, the long-tailed distribution of multilingual corpora poses a challenge of Pareto optimization, i.e., optimizing for some languages may come at the cos
Externí odkaz:
http://arxiv.org/abs/2305.15718
Although all-in-one-model multilingual neural machine translation (multilingual NMT) has achieved remarkable progress, the convergence inconsistency in the joint training is ignored, i.e., different language pairs reaching convergence in different ep
Externí odkaz:
http://arxiv.org/abs/2205.01620
Recently, various neural encoder-decoder models pioneered by Seq2Seq framework have been proposed to achieve the goal of generating more abstractive summaries by learning to map input text to output text. At a high level, such neural models can freel
Externí odkaz:
http://arxiv.org/abs/2104.14839
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.