• 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.
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
TLDR
BART is presented, a denoising autoencoder for pretraining sequence-to-sequence models, which matches the performance of RoBERTa on GLUE and SQuAD, and achieves new state-of-the-art results on a range of abstractive dialogue, question answering, and summarization tasks.
XNLI: Evaluating Cross-lingual Sentence Representations
TLDR
This work constructs an evaluation set for XLU by extending the development and test sets of the Multi-Genre Natural Language Inference Corpus to 14 languages, including low-resource languages such as Swahili and Urdu and finds that XNLI represents a practical and challenging evaluation suite and that directly translating the test data yields the best performance among available baselines.
SemEval-2013 Task 2: Sentiment Analysis in Twitter
TLDR
Crowdourcing on Amazon Mechanical Turk was used to label a large Twitter training dataset along with additional test sets of Twitter and SMS messages for both subtasks, which included two subtasks: A, an expression-level subtask, and B, a message level subtask.
SemEval-2016 Task 4: Sentiment Analysis in Twitter
TLDR
The fourth year of the SemEval-2016 Task 4 comprises five subtasks, three of which represent a significant departure from previous editions, and the task continues to be very popular, attracting a total of 43 teams.
SemEval-2015 Task 10: Sentiment Analysis in Twitter
TLDR
The 2015 iteration of the SemEval shared task on Sentiment Analysis in Twitter was the most popular sentiment analysis shared task to date with more than 40 teams participating in each of the last three years.
Emerging Cross-lingual Structure in Pretrained Language Models
TLDR
It is shown that transfer is possible even when there is no shared vocabulary across the monolingual corpora and also when the text comes from very different domains, and it is strongly suggested that, much like for non-contextual word embeddings, there are universal latent symmetries in the learned embedding spaces.
SemEval-2014 Task 9: Sentiment Analysis in Twitter
TLDR
The Sentiment Analysis in Twitter task is described, a continuation of the last year’s task that ran successfully as part of SemEval2013 and introduced three new test sets: regular tweets, sarcastic tweets, and LiveJournal sentences.
Conundrums in Noun Phrase Coreference Resolution: Making Sense of the State-of-the-Art
TLDR
This work examines three subproblems that play a role in coreference resolution: named entity recognition, anaphoricity determination, and coreference element detection, and measures the performance of a state-of-the-art coreference resolver on several classes of anaphora.
...
1
2
3
4
5
...