Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Naomi Saphra

Naomi Saphra

The MultiBERTs: BERT Reproductions for Robustness Analysis


Jun 30, 2021
Thibault Sellam, Steve Yadlowsky, Jason Wei, Naomi Saphra, Alexander D'Amour, Tal Linzen, Jasmijn Bastings, Iulia Turc, Jacob Eisenstein, Dipanjan Das, Ian Tenney, Ellie Pavlick

Figure 1 for The MultiBERTs: BERT Reproductions for Robustness Analysis
Figure 2 for The MultiBERTs: BERT Reproductions for Robustness Analysis
Figure 3 for The MultiBERTs: BERT Reproductions for Robustness Analysis
Figure 4 for The MultiBERTs: BERT Reproductions for Robustness Analysis
* Checkpoints and example analyses: http://goo.gle/multiberts 
  Access Paper or Ask Questions

A Non-Linear Structural Probe


May 21, 2021
Jennifer C. White, Tiago Pimentel, Naomi Saphra, Ryan Cotterell

Figure 1 for A Non-Linear Structural Probe
* Accepted at NAACL 2021 
  Access Paper or Ask Questions

LSTMs Compose (and Learn) Bottom-Up


Oct 06, 2020
Naomi Saphra, Adam Lopez

Figure 1 for LSTMs Compose (and Learn) Bottom-Up
Figure 2 for LSTMs Compose (and Learn) Bottom-Up
Figure 3 for LSTMs Compose (and Learn) Bottom-Up
Figure 4 for LSTMs Compose (and Learn) Bottom-Up
* Published in EMNLP Findings 2020. arXiv admin note: substantial text overlap with arXiv:2004.13195 
  Access Paper or Ask Questions

Pareto Probing: Trading Off Accuracy for Complexity


Oct 05, 2020
Tiago Pimentel, Naomi Saphra, Adina Williams, Ryan Cotterell

Figure 1 for Pareto Probing: Trading Off Accuracy for Complexity
Figure 2 for Pareto Probing: Trading Off Accuracy for Complexity
Figure 3 for Pareto Probing: Trading Off Accuracy for Complexity
Figure 4 for Pareto Probing: Trading Off Accuracy for Complexity
* Tiago Pimentel and Naomi Saphra contributed equally to this work. Accepted for publication at EMNLP 2020. Code available in https://github.com/rycolab/pareto-probing 
  Access Paper or Ask Questions

Word Interdependence Exposes How LSTMs Compose Representations


Apr 27, 2020
Naomi Saphra, Adam Lopez

Figure 1 for Word Interdependence Exposes How LSTMs Compose Representations
Figure 2 for Word Interdependence Exposes How LSTMs Compose Representations
Figure 3 for Word Interdependence Exposes How LSTMs Compose Representations
Figure 4 for Word Interdependence Exposes How LSTMs Compose Representations

  Access Paper or Ask Questions

How to Evaluate Word Representations of Informal Domain?


Nov 13, 2019
Yekun Chai, Naomi Saphra, Adam Lopez

Figure 1 for How to Evaluate Word Representations of Informal Domain?
Figure 2 for How to Evaluate Word Representations of Informal Domain?
Figure 3 for How to Evaluate Word Representations of Informal Domain?
Figure 4 for How to Evaluate Word Representations of Informal Domain?

  Access Paper or Ask Questions

Sparsity Emerges Naturally in Neural Language Models


Jul 22, 2019
Naomi Saphra, Adam Lopez

Figure 1 for Sparsity Emerges Naturally in Neural Language Models
Figure 2 for Sparsity Emerges Naturally in Neural Language Models
Figure 3 for Sparsity Emerges Naturally in Neural Language Models
Figure 4 for Sparsity Emerges Naturally in Neural Language Models
* Published in the ICML 2019 Workshop on Identifying and Understanding Deep Learning Phenomena: https://openreview.net/forum?id=H1ets1h56E 
  Access Paper or Ask Questions

Understanding Learning Dynamics Of Language Models with SVCCA


Nov 01, 2018
Naomi Saphra, Adam Lopez

Figure 1 for Understanding Learning Dynamics Of Language Models with SVCCA
Figure 2 for Understanding Learning Dynamics Of Language Models with SVCCA
Figure 3 for Understanding Learning Dynamics Of Language Models with SVCCA
Figure 4 for Understanding Learning Dynamics Of Language Models with SVCCA

  Access Paper or Ask Questions

DyNet: The Dynamic Neural Network Toolkit


Jan 15, 2017
Graham Neubig, Chris Dyer, Yoav Goldberg, Austin Matthews, Waleed Ammar, Antonios Anastasopoulos, Miguel Ballesteros, David Chiang, Daniel Clothiaux, Trevor Cohn, Kevin Duh, Manaal Faruqui, Cynthia Gan, Dan Garrette, Yangfeng Ji, Lingpeng Kong, Adhiguna Kuncoro, Gaurav Kumar, Chaitanya Malaviya, Paul Michel, Yusuke Oda, Matthew Richardson, Naomi Saphra, Swabha Swayamdipta, Pengcheng Yin

Figure 1 for DyNet: The Dynamic Neural Network Toolkit
Figure 2 for DyNet: The Dynamic Neural Network Toolkit
Figure 3 for DyNet: The Dynamic Neural Network Toolkit
Figure 4 for DyNet: The Dynamic Neural Network Toolkit
* 33 pages 
  Access Paper or Ask Questions

Evaluating Informal-Domain Word Representations With UrbanDictionary


Jun 27, 2016
Naomi Saphra, Adam Lopez


  Access Paper or Ask Questions

A framework for (under)specifying dependency syntax without overloading annotators


Jun 15, 2013
Nathan Schneider, Brendan O'Connor, Naomi Saphra, David Bamman, Manaal Faruqui, Noah A. Smith, Chris Dyer, Jason Baldridge

Figure 1 for A framework for (under)specifying dependency syntax without overloading annotators
Figure 2 for A framework for (under)specifying dependency syntax without overloading annotators
Figure 3 for A framework for (under)specifying dependency syntax without overloading annotators
Figure 4 for A framework for (under)specifying dependency syntax without overloading annotators
* This is an expanded version of a paper appearing in Proceedings of the 7th Linguistic Annotation Workshop & Interoperability with Discourse, Sofia, Bulgaria, August 8-9, 2013 
  Access Paper or Ask Questions