• Publications
  • Influence
RoBERTa: A Robustly Optimized BERT Pretraining Approach
TLDR
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.
Unsupervised Cross-lingual Representation Learning at Scale
TLDR
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.
Finding Deceptive Opinion Spam by Any Stretch of the Imagination
TLDR
This work develops and compares three approaches to detecting deceptive opinion spam, and develops a classifier that is nearly 90% accurate on the authors' gold-standard opinion spam dataset, and reveals a relationship between deceptive opinions and imaginative writing.
fairseq: A Fast, Extensible Toolkit for Sequence Modeling
TLDR
Fairseq is an open-source sequence modeling toolkit that allows researchers and developers to train custom models for translation, summarization, language modeling, and other text generation tasks and supports distributed training across multiple GPUs and machines.
Phrase-Based & Neural Unsupervised Machine Translation
TLDR
This work investigates how to learn to translate when having access to only large monolingual corpora in each language, and proposes two model variants, a neural and a phrase-based model, which are significantly better than methods from the literature, while being simpler and having fewer hyper-parameters.
Understanding Back-Translation at Scale
TLDR
This work broadens the understanding of back-translation and investigates a number of methods to generate synthetic source sentences, finding that in all but resource poor settings back-translations obtained via sampling or noised beam outputs are most effective.
Scaling Neural Machine Translation
TLDR
This paper shows that reduced precision and large batch training can speedup training by nearly 5x on a single 8-GPU machine with careful tuning and implementation.
Negative Deceptive Opinion Spam
TLDR
This work creates and study the first dataset of deceptive opinion spam with negative sentiment reviews, and finds that standard n-gram text categorization techniques can detect negative deceptive opinions spam with performance far surpassing that of human judges.
The FLORES Evaluation Datasets for Low-Resource Machine Translation: Nepali–English and Sinhala–English
TLDR
This work introduces the FLORES evaluation datasets for Nepali–English and Sinhala– English, based on sentences translated from Wikipedia, and demonstrates that current state-of-the-art methods perform rather poorly on this benchmark, posing a challenge to the research community working on low-resource MT.
Recipes for Building an Open-Domain Chatbot
TLDR
Human evaluations show the best models outperform existing approaches in multi-turn dialogue on engagingness and humanness measurements, and the limitations of this work are discussed by analyzing failure cases of the models.
...
1
2
3
4
5
...