中文
相关论文

相关论文: A Systematic Analysis of Subwords and Cross-Lingua…

200 篇论文

Cross-lingual representations of words enable us to reason about word meaning in multilingual contexts and are a key facilitator of cross-lingual transfer when developing natural language processing models for low-resource languages. In…

计算与语言 · 计算机科学 2019-10-08 Sebastian Ruder , Ivan Vulić , Anders Søgaard

Multilingual machine translation systems aim to make knowledge accessible across languages, yet learning effective cross-lingual representations remains challenging. These challenges are especially pronounced for low-resource languages,…

计算与语言 · 计算机科学 2026-01-08 David Stap

Large multilingual language models typically share their parameters across all languages, which enables cross-lingual task transfer, but learning can also be hindered when training updates from different languages are in conflict. In this…

计算与语言 · 计算机科学 2022-11-02 Rochelle Choenni , Dan Garrette , Ekaterina Shutova

Cross-lingual transfer of word embeddings aims to establish the semantic mappings among words in different languages by learning the transformation functions over the corresponding word embedding spaces. Successfully solving this problem…

计算与语言 · 计算机科学 2018-09-12 Ruochen Xu , Yiming Yang , Naoki Otani , Yuexin Wu

Subword regularizations use multiple subword segmentations during training to improve the robustness of neural machine translation models. In previous subword regularizations, we use multiple segmentations in the training process but use…

计算与语言 · 计算机科学 2022-03-28 Sho Takase , Tatsuya Hiraoka , Naoaki Okazaki

Cross-lingual model transfer is a compelling and popular method for predicting annotations in a low-resource language, whereby parallel corpora provide a bridge to a high-resource language and its associated annotated corpora. However,…

计算与语言 · 计算机科学 2017-05-02 Meng Fang , Trevor Cohn

Subword regularization, used widely in NLP, improves model performance by reducing the dependency on exact tokenizations, augmenting the training corpus, and exposing the model to more unique contexts during training. BPE and MaxMatch, two…

计算与语言 · 计算机科学 2024-08-22 Marco Cognetta , Vilém Zouhar , Naoaki Okazaki

Subword tokenizers trained on multilingual corpora naturally produce overlapping tokens across languages. Does token overlap facilitate cross-lingual transfer or instead introduce interference between languages? Prior work offers mixed…

计算与语言 · 计算机科学 2025-09-26 Julie Kallini , Dan Jurafsky , Christopher Potts , Martijn Bartelds

Subword segmenters like BPE operate as a preprocessing step in neural machine translation and other (conditional) language models. They are applied to datasets before training, so translation or text generation quality relies on the quality…

计算与语言 · 计算机科学 2023-05-12 Francois Meyer , Jan Buys

Cross-lingual vocabulary transfer plays a promising role in adapting pre-trained language models to new languages, including low-resource languages. Existing approaches that utilize monolingual or parallel corpora face challenges when…

计算与语言 · 计算机科学 2025-06-03 Haruki Sakajo , Yusuke Ide , Justin Vasselli , Yusuke Sakai , Yingtao Tian , Hidetaka Kamigaito , Taro Watanabe

Multilingual pretrained representations generally rely on subword segmentation algorithms to create a shared multilingual vocabulary. However, standard heuristic algorithms often lead to sub-optimal segmentation, especially for languages…

计算与语言 · 计算机科学 2021-04-07 Xinyi Wang , Sebastian Ruder , Graham Neubig

Subword units are an effective way to alleviate the open vocabulary problems in neural machine translation (NMT). While sentences are usually converted into unique subword sequences, subword segmentation is potentially ambiguous and…

计算与语言 · 计算机科学 2018-05-01 Taku Kudo

Multilingual pre-trained language models transfer remarkably well on cross-lingual downstream tasks. However, the extent to which they learn language-neutral representations (i.e., shared representations that encode similar phenomena across…

计算与语言 · 计算机科学 2022-11-01 Negar Foroutan , Mohammadreza Banaei , Remi Lebret , Antoine Bosselut , Karl Aberer

Subword modeling for zero-resource languages aims to learn low-level representations of speech audio without using transcriptions or other resources from the target language (such as text corpora or pronunciation dictionaries). A good…

音频与语音处理 · 电气工程与系统科学 2020-04-20 Enno Hermann , Herman Kamper , Sharon Goldwater

Multilingual pretraining and fine-tuning have remarkably succeeded in various natural language processing tasks. Transferring representations from one language to another is especially crucial for cross-lingual learning. One can expect…

计算与语言 · 计算机科学 2024-03-26 Shaoxiong Ji , Timothee Mickus , Vincent Segonne , Jörg Tiedemann

Transfer learning or multilingual model is essential for low-resource neural machine translation (NMT), but the applicability is limited to cognate languages by sharing their vocabularies. This paper shows effective techniques to transfer a…

计算与语言 · 计算机科学 2019-06-06 Yunsu Kim , Yingbo Gao , Hermann Ney

Using a vocabulary that is shared across languages is common practice in Multilingual Neural Machine Translation (MNMT). In addition to its simple design, shared tokens play an important role in positive knowledge transfer, assuming that…

计算与语言 · 计算机科学 2024-01-23 Di Wu , Christof Monz

Subwords are the most widely used output units in end-to-end speech recognition. They combine the best of two worlds by modeling the majority of frequent words directly and at the same time allow open vocabulary speech recognition by…

音频与语音处理 · 电气工程与系统科学 2020-08-11 Egor Lakomkin , Jahn Heymann , Ilya Sklyar , Simon Wiesler

Most neural machine translation systems are built upon subword units extracted by methods such as Byte-Pair Encoding (BPE) or wordpiece. However, the choice of number of merge operations is generally made by following existing recipes. In…

计算与语言 · 计算机科学 2019-06-26 Shuoyang Ding , Adithya Renduchintala , Kevin Duh

Cross-lingual word embeddings aim to bridge the gap between high-resource and low-resource languages by allowing to learn multilingual word representations even without using any direct bilingual signal. The lion's share of the methods are…

计算与语言 · 计算机科学 2020-09-03 Magdalena Biesialska , Marta R. Costa-jussà
‹ 上一页 1 2 3 10 下一页 ›