Structural Biases for Improving Transformers on Translation into Morphologically Rich Languages MTSummit

Paul Soulos | Sudha Rao | Caitlin Smith | Eric Rosen | Asli Celikyilmaz | R. Thomas McCoy | Yichen Jiang | Coleman Haley | Roland Fernandez | Hamid Palangi | Jianfeng Gao | Paul Smolensky |

Syntactic Data Augmentation Increases Robustness to Inference Heuristics ACL

Junghyun Min | R. Thomas McCoy | Dipanjan Das | Emily Pitler | Tal Linzen |

Representations of Syntax [MASK] Useful: Effects of Constituency and Dependency Structure in Recursive LSTMs ACL

Michael Lepori | Tal Linzen | R. Thomas McCoy |

Picking BERT’s Brain: Probing for Linguistic Dependencies in Contextualized Embeddings Using Representational Similarity Analysis COLING

Michael Lepori | R. Thomas McCoy |

Does Syntax Need to Grow on Trees? Sources of Hierarchical Inductive Bias in Sequence-to-Sequence Networks TACL

R. Thomas McCoy | Robert Frank | Tal Linzen |

Discovering the Compositional Structure of Vector Representations with Role Learning Networks BlackboxNLP EMNLP

Paul Soulos | R. Thomas McCoy | Tal Linzen | Paul Smolensky |

BERTs of a feather do not generalize together: Large variability in generalization across models with similar test set performance BlackboxNLP EMNLP

R. Thomas McCoy | Junghyun Min | Tal Linzen |

Can You Tell Me How to Get Past Sesame Street? Sentence-Level Pretraining Beyond Language Modeling ACL

Alex Wang | Jan Hula | Patrick Xia | Raghavendra Pappagari | R. Thomas McCoy | Roma Patel | Najoung Kim | Ian Tenney | Yinghui Huang | Katherin Yu | Shuning Jin | Berlin Chen | Benjamin Van Durme | Edouard Grave | Ellie Pavlick | Samuel R. Bowman |

Linguistically Rich Vector Representations of Supertags for TAG Parsing TAG+ WS

Dan Friedman | Jungo Kasai | R. Thomas McCoy | Robert Frank | Forrest Davis | Owen Rambow |