• Publications
  • Influence
RoBERTa: A Robustly Optimized BERT Pretraining Approach
TLDR
We present a replication study of BERT pretraining (Devlin et al., 2019) that carefully measures the impact of many key hyperparameters and training data size. Expand
  • 2,650
  • 839
  • PDF
Unsupervised Cross-lingual Representation Learning at Scale
TLDR
We present XLM-R a transformer-based multilingual masked language model pre-trained on one hundred languages, which obtains state-of-the-art performance on cross-lingual classification, sequence labeling and question answering. Expand
  • 490
  • 193
  • PDF
BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension
TLDR
We present BART, a denoising autoencoder for pretraining sequence-to-sequence models. Expand
  • 553
  • 141
  • PDF
XNLI: Evaluating Cross-lingual Sentence Representations
TLDR
We extend the Multi-Genre Natural Language Inference Corpus (MultiNLI) to 14 languages, including low-resource languages such as Swahili and Urdu. Expand
  • 285
  • 90
  • PDF
SemEval-2013 Task 2: Sentiment Analysis in Twitter
TLDR
We have proposed SemEval-2013 Task 2: Sentiment Analysis in Twitter, which included two subtasks: A, an expression-level subtask, and B, a message level subtask. Expand
  • 624
  • 82
  • PDF
SemEval-2017 Task 4: Sentiment Analysis in Twitter
TLDR
This paper discusses the fourth year of the ”Sentiment Analysis in Twitter Task” task at SemEval and introduces a new language, Arabic. Expand
  • 527
  • 72
  • PDF
SemEval-2015 Task 10: Sentiment Analysis in Twitter
TLDR
In this paper, we describe the 2015 iteration of the SemEval shared task on Sentiment Analysis in Twitter. Expand
  • 363
  • 37
  • PDF
SemEval-2014 Task 9: Sentiment Analysis in Twitter
TLDR
We describe the Sentiment Analysis in Twitter task, ran as part of SemEval-2014. Expand
  • 287
  • 33
  • PDF
Conundrums in Noun Phrase Coreference Resolution: Making Sense of the State-of-the-Art
TLDR
We aim to shed light on the state-of-the-art in NP coreference resolution by teasing apart the differences in the MUC and ACE task definitions, the assumptions made in evaluation methodologies, and inherent differences in text corpora. Expand
  • 152
  • 21
  • PDF
Evaluation Measures for the SemEval-2016 Task 4 “Sentiment Analysis in Twitter” (Draft: Version 1.13)
This informal document details the evaluation measures that will be used in SemEval2016 Task 4 “Sentiment Analysis in Twitter”, a revamped edition of SemEval-2015 Task 10 (Rosenthal et al., 2015).Expand
  • 138
  • 20
  • PDF
...
1
2
3
4
5
...