activity
20102026
most citedTransition-Based Dependency Parsing with Stack Long Short-Term Memory

526 citations · 1.4k across the 46 of their papers we have counts for

collaborators
Showing 2021Show all

13 papers · 1 filter

cs.CL2021

Expected Validation Performance and Estimation of a Random Variable's Maximum

Jesse Dodge, Suchin Gururangan, Dallas Card +2

Research in NLP is often supported by experimental results, and improved reporting of such results can lead to better understanding and more reproducible science. In this paper we…

cs.CL2021

Sentence Bottleneck Autoencoders from Transformer Language Models

Ivan Montero, Nikolaos Pappas, Noah A. Smith

Representation learning for text via pretraining a language model on a large corpus has become a standard starting point for building NLP systems. This approach stands in contrast…

cs.CL2021

DEMix Layers: Disentangling Domains for Modular Language Modeling

Suchin Gururangan, Mike Lewis, Ari Holtzman +2

We introduce a new domain expert mixture (DEMix) layer that enables conditioning a language model (LM) on the domain of the input text. A DEMix layer is a collection of expert feed…

cs.CL2021★ 16 cited

All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Text

Elizabeth Clark, Tal August, Sofia Serrano +3

Human evaluations are typically considered the gold standard in natural language generation, but as models' fluency improves, how well can evaluators detect and judge machine-gener…

cs.CL2021★ 11 cited

DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts

Alisa Liu, Maarten Sap, Ximing Lu +4

Despite recent advances in natural language generation, it remains challenging to control attributes of generated text. We propose DExperts: Decoding-time Experts, a decoding-time…

cs.CL2021★ 13 cited

Scientific Language Models for Biomedical Knowledge Base Completion: An Empirical Study

Rahul Nadkarni, David Wadden, Iz Beltagy +3

Biomedical knowledge graphs (KGs) hold rich information on entities such as diseases, drugs, and genes. Predicting missing links in these graphs can boost many important applicatio…