• Publications
  • Influence
SWAG: A Large-Scale Adversarial Dataset for Grounded Commonsense Inference
TLDR
We introduce Adversarial Filtering (AF), a novel procedure that constructs a de-biased dataset by iteratively training an ensemble of stylistic classifiers and using them to filter the data. Expand
  • 236
  • 40
  • PDF
Defending Against Neural Fake News
TLDR
We present a new generative model for controllable text generation called Grover. Expand
  • 160
  • 32
  • PDF
From Recognition to Cognition: Visual Commonsense Reasoning
TLDR
We present a new reasoning engine, Recognition to Cognition Networks (R2C), that models the necessary layered inferences for grounding, contextualization, and reasoning. Expand
  • 134
  • 29
  • PDF
Synthetic and Natural Noise Both Break Neural Machine Translation
TLDR
Character-based neural machine translation (NMT) models alleviate out-of-vocabulary issues, learn morphology, and move us closer to completely end-to-end translation systems. Expand
  • 260
  • 27
  • PDF
Supertagging With LSTMs
TLDR
We introduce a new bi–LSTM model that achieves state-of-the-art performance on CCG supertagging and parsing, and a detailed analysis of the quality of various LSTM architectures, forward, backward, and bi-directional. Expand
  • 55
  • 14
  • PDF
HellaSwag: Can a Machine Really Finish Your Sentence?
TLDR
We show that commonsense inference still proves difficult for even state-of-the-art models, by presenting HellaSwag, a new challenge dataset. Expand
  • 84
  • 11
  • PDF
Natural Language Communication with Robots
TLDR
We propose a framework for devising empirically testable algorithms for bridging the communication gap between humans and robots, in which humans give instructions to robots using unrestricted natural language commands. Expand
  • 62
  • 11
  • PDF
Tactical Rewind: Self-Correction via Backtracking in Vision-And-Language Navigation
TLDR
We present the Frontier Aware Search with backTracking (FAST) Navigator, a general framework for action decoding, that achieves state-of-the-art results on the 2018 Room-to-Room (R2R) Vision-and-Language navigation challenge. Expand
  • 44
  • 9
  • PDF
Unsupervised Neural Hidden Markov Models
TLDR
In this work, we present the first results for neuralizing an Unsupervised Hidden Markov Model using Expectation-Maximization and Variational Inference. Expand
  • 31
  • 6
  • PDF
ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks
TLDR
We present ALFRED (Action Learning From Realistic Environments and Directives), a benchmark for learning a mapping from natural language instructions and egocentric vision to sequences of actions for household tasks. Expand
  • 28
  • 5
  • PDF