mT5: A Massively Multilingual Pre-trained Text-to-Text Transformer
Linting Xue,Noah Constant,Adam Roberts,Mihir Kale,Rami Al-Rfou,Aditya Siddhant,Aditya Barua,Colin Raffel +7 more
- pp 483-498
Reads0
Chats0
TLDR
This paper proposed a multilingual variant of T5, mT5, which was pre-trained on a new Common Crawl-based dataset covering 101 languages and achieved state-of-the-art performance on many multilingual benchmarks.Abstract:
The recent “Text-to-Text Transfer Transformer” (T5) leveraged a unified text-to-text format and scale to attain state-of-the-art results on a wide variety of English-language NLP tasks. In this paper, we introduce mT5, a multilingual variant of T5 that was pre-trained on a new Common Crawl-based dataset covering 101 languages. We detail the design and modified training of mT5 and demonstrate its state-of-the-art performance on many multilingual benchmarks. We also describe a simple technique to prevent “accidental translation” in the zero-shot setting, where a generative model chooses to (partially) translate its prediction into the wrong language. All of the code and model checkpoints used in this work are publicly available.read more
Citations
More filters
Proceedings Article
ALEXSIS: A Dataset for Lexical Simplification in Spanish
Daniel Ferrés,Horacio Saggion +1 more
TL;DR: In this paper , the authors introduce a new dataset for lexical simplification, ALEXSIS, and use it to benchmark the state-of-the-art systems in Spanish.
Journal ArticleDOI
Cross-Lingual Dialogue Dataset Creation via Outline-Based Generation
TL;DR: This work proposes a novel outline-based annotation process for multilingual TOD datasets, where domainspecific abstract schemata of dialogue are mapped into natural language outlines, and enables natural language understanding, dialogue state tracking, and end-toend dialogue modelling and evaluation in 4 diverse languages.
Journal ArticleDOI
Quantifying and alleviating political bias in language models
TL;DR: This paper proposed a reinforcement learning framework for mitigating political bias in generated text by using rewards from word embeddings or a classifier, which can guide debiased generation without having access to the training data or requiring the model to be retrained.
Proceedings ArticleDOI
LiT: Zero-Shot Transfer with Locked-image text Tuning
TL;DR: LiT as mentioned in this paper aligns image and text mod-els while still taking advantage of their pre-training using contrastive training to achieve zero-shot transfer to new vision tasks, such as image classification or retrieval.
Journal ArticleDOI
PreSTU: Pre-Training for Scene-Text Understanding
Jihyung Kil,Soravit Changpinyo,Xi Chen,Hexiang Hu,Sebastian Goodman,Wei-Lun Chao,Radu Soricut +6 more
TL;DR: PreSTU as discussed by the authors introduces OCR-aware pre-training objectives that encourage the model to recognize text from an image and to connect what is recognized to the rest of the image content.
References
More filters
Proceedings Article
Attention is All you Need
Ashish Vaswani,Noam Shazeer,Niki Parmar,Jakob Uszkoreit,Llion Jones,Aidan N. Gomez,Lukasz Kaiser,Illia Polosukhin +7 more
TL;DR: This paper proposed a simple network architecture based solely on an attention mechanism, dispensing with recurrence and convolutions entirely and achieved state-of-the-art performance on English-to-French translation.
Posted Content
RoBERTa: A Robustly Optimized BERT Pretraining Approach
Yinhan Liu,Myle Ott,Naman Goyal,Jingfei Du,Mandar Joshi,Danqi Chen,Omer Levy,Michael Lewis,Luke Zettlemoyer,Veselin Stoyanov +9 more
TL;DR: It is found that BERT was significantly undertrained, and can match or exceed the performance of every model published after it, and the best model achieves state-of-the-art results on GLUE, RACE and SQuAD.
Proceedings ArticleDOI
SQuAD: 100,000+ Questions for Machine Comprehension of Text
TL;DR: The Stanford Question Answering Dataset (SQuAD) as mentioned in this paper is a reading comprehension dataset consisting of 100,000+ questions posed by crowdworkers on a set of Wikipedia articles, where the answer to each question is a segment of text from the corresponding reading passage.
Proceedings ArticleDOI
Unsupervised Cross-lingual Representation Learning at Scale
Alexis Conneau,Kartikay Khandelwal,Naman Goyal,Vishrav Chaudhary,Guillaume Wenzek,Francisco Guzmán,Edouard Grave,Myle Ott,Luke Zettlemoyer,Veselin Stoyanov +9 more
TL;DR: It is shown that pretraining multilingual language models at scale leads to significant performance gains for a wide range of cross-lingual transfer tasks, and the possibility of multilingual modeling without sacrificing per-language performance is shown for the first time.
Proceedings ArticleDOI
Universal Language Model Fine-tuning for Text Classification
Jeremy Howard,Sebastian Ruder +1 more
TL;DR: Universal Language Model Fine-tuning (ULMFiT) as mentioned in this paper is an effective transfer learning method that can be applied to any task in NLP, and introduces techniques that are key for finetuning a language model.