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 Michael Hahn

Michael Hahn

Sensitivity as a Complexity Measure for Sequence Classification Tasks


Apr 21, 2021
Michael Hahn, Dan Jurafsky, Richard Futrell

Figure 1 for Sensitivity as a Complexity Measure for Sequence Classification Tasks
Figure 2 for Sensitivity as a Complexity Measure for Sequence Classification Tasks
Figure 3 for Sensitivity as a Complexity Measure for Sequence Classification Tasks
Figure 4 for Sensitivity as a Complexity Measure for Sequence Classification Tasks
* Accepted by TACL. This is a pre-MIT Press publication version 
  Access Paper or Ask Questions

RNNs can generate bounded hierarchical languages with optimal memory


Oct 15, 2020
John Hewitt, Michael Hahn, Surya Ganguli, Percy Liang, Christopher D. Manning

Figure 1 for RNNs can generate bounded hierarchical languages with optimal memory
Figure 2 for RNNs can generate bounded hierarchical languages with optimal memory
Figure 3 for RNNs can generate bounded hierarchical languages with optimal memory
Figure 4 for RNNs can generate bounded hierarchical languages with optimal memory
* EMNLP2020 + appendix typo fixes 
  Access Paper or Ask Questions

Tabula nearly rasa: Probing the Linguistic Knowledge of Character-Level Neural Language Models Trained on Unsegmented Text


Jun 17, 2019
Michael Hahn, Marco Baroni

* Accepted by Transactions of the Association for Computational Linguistics 
  Access Paper or Ask Questions

Theoretical Limitations of Self-Attention in Neural Sequence Models


Jun 16, 2019
Michael Hahn

Figure 1 for Theoretical Limitations of Self-Attention in Neural Sequence Models

  Access Paper or Ask Questions

Character-based Surprisal as a Model of Human Reading in the Presence of Errors


Feb 02, 2019
Michael Hahn, Frank Keller, Yonatan Bisk, Yonatan Belinkov

Figure 1 for Character-based Surprisal as a Model of Human Reading in the Presence of Errors
Figure 2 for Character-based Surprisal as a Model of Human Reading in the Presence of Errors
Figure 3 for Character-based Surprisal as a Model of Human Reading in the Presence of Errors
Figure 4 for Character-based Surprisal as a Model of Human Reading in the Presence of Errors
* Under double-blind peer review, please do not disseminate widely 
  Access Paper or Ask Questions

Modeling Task Effects in Human Reading with Neural Attention


Nov 02, 2018
Michael Hahn, Frank Keller

Figure 1 for Modeling Task Effects in Human Reading with Neural Attention
Figure 2 for Modeling Task Effects in Human Reading with Neural Attention
Figure 3 for Modeling Task Effects in Human Reading with Neural Attention
Figure 4 for Modeling Task Effects in Human Reading with Neural Attention

  Access Paper or Ask Questions

Modeling Human Reading with Neural Attention


Apr 24, 2017
Michael Hahn, Frank Keller

Figure 1 for Modeling Human Reading with Neural Attention
Figure 2 for Modeling Human Reading with Neural Attention
Figure 3 for Modeling Human Reading with Neural Attention
Figure 4 for Modeling Human Reading with Neural Attention
* EMNLP 2016, pp. 85-95, Austin, TX 
  Access Paper or Ask Questions