• Publications
  • Influence
Reformer: The Efficient Transformer
TLDR
Large Transformer models routinely achieve state-of-the-art results on a number of tasks but training these models can be prohibitively costly, especially on long sequences. Expand
Constituency Parsing with a Self-Attentive Encoder
TLDR
We demonstrate that replacing an LSTM encoder with a self-attentive architecture can lead to improvements to a state-of-the-art discriminative constituency parser. Expand
BOSS: Building Operating System Services
TLDR
We develop a set of operating system services called BOSS, which supports multiple portable, fault-tolerant applications on top of the distributed physical resources present in large commercial buildings. Expand
Multilingual Alignment of Contextual Word Representations
TLDR
We propose procedures for evaluating and strengthening contextual embedding alignment and show that they are useful in analyzing and improving multilingual BERT. Expand
Multilingual Constituency Parsing with Self-Attention and Pre-Training
TLDR
We show that joint multilingual pre-training and fine-tuning allows sharing all but a small number of parameters between ten languages in the final model. Expand
KERMIT: Generative Insertion-Based Modeling for Sequences
TLDR
We present KERMIT, a simple insertion-based approach to generative modeling for sequences and sequence pairs using a single neural network and, unlike much prior work, does not rely on a prespecified factorization of the data distribution. Expand
Physics-based trajectory optimization for grasping in cluttered environments
TLDR
In this work, we present a physics-based trajectory optimization approach for planning grasp approach trajectories for grasping in cluttered environments. Expand
Building application stack (BAS)
TLDR
We present BAS, an application programming interface and runtime that enables writing portable code by providing methods to explicitly handle differences in building designs. Expand
CoDraw: Collaborative Drawing as a Testbed for Grounded Goal-driven Communication
TLDR
In this work, we propose a goal-driven collaborative task that combines language, perception, and action. Expand
Cross-Domain Generalization of Neural Constituency Parsers
TLDR
We present three results about the generalization of neural parsers in a zero-shot setting: training on trees from one corpus and evaluating on out-of-domain corpora. Expand
...
1
2
...