Research - Papers
Explore a selection of our published work on a variety of key research challenges in AI.
Quoref: A Reading Comprehension Dataset with Questions Requiring Coreferential Reasoning
Machine comprehension of texts longer than a single sentence often requires coreference resolution. However, most current reading comprehension benchmarks do not contain complex coreferential…
RNN Architecture Learning with Sparse Regularization
Neural models for NLP typically use large numbers of parameters to reach state-of-the-art performance, which can lead to excessive memory usage and increased runtime. We present a structure learning…
SciBERT: A Pretrained Language Model for Scientific Text
Obtaining large-scale annotated data for NLP tasks in the scientific domain is challenging and expensive. We release SciBERT, a pretrained language model based on BERT (Devlin et al., 2018) to…
Show Your Work: Improved Reporting of Experimental Results
Research in natural language processing proceeds, in part, by demonstrating that new models achieve superior performance (e.g., accuracy) on held-out test data, compared to previous results. In this…
Social IQA: Commonsense Reasoning about Social Interactions
We introduce Social IQa, the first largescale benchmark for commonsense reasoning about social situations. Social IQa contains 38,000 multiple choice questions for probing emotional and social…
SpanBERT: Improving Pre-training by Representing and Predicting Spans
We present SpanBERT, a pre-training method that is designed to better represent and predict spans of text. Our approach extends BERT by (1) masking contiguous random spans, rather than random…
Topics to Avoid: Demoting Latent Confounds in Text Classification
Despite impressive performance on many text classification tasks, deep neural networks tend to learn frequent superficial patterns that are specific to the training data and do not always generalize…
Transfer Learning Between Related Tasks Using Expected Label Proportions
Deep learning systems thrive on abundance of labeled training data but such data is not always available, calling for alternative methods of supervision. One such method is expectation…
Universal Adversarial Triggers for Attacking and Analyzing NLP
dversarial examples highlight model vulnerabilities and are useful for evaluation and interpretation. We define universal adversarial triggers: input-agnostic sequences of tokens that trigger a…
WIQA: A dataset for "What if..." reasoning over procedural text
We introduce WIQA, the first large-scale dataset of "What if..." questions over procedural text. WIQA contains three parts: a collection of paragraphs each describing a process, e.g., beach erosion;…