Strong generative capacity of morphological processes

no code implementations • SCiL 2021 • Hossep Dolatian, Jonathan Rawski, Jeffrey Heinz

Probing RNN Encoder-Decoder Generalization of Subregular Functions using Reduplication

no code implementations • SCiL 2020 • Max Nelson, Hossep Dolatian, Jonathan Rawski, Brandon Prickett

Multi-Input Strictly Local Functions for Templatic Morphology

no code implementations • SCiL 2020 • Hossep Dolatian, Jonathan Rawski

Multi-Input Strictly Local Functions for Tonal Phonology

no code implementations • SCiL 2020 • Jonathan Rawski, Hossep Dolatian

Transformers as Transducers

no code implementations • 2 Apr 2024 • Lena Strobl, Dana Angluin, David Chiang, Jonathan Rawski, Ashish Sabharwal

We study the sequence-to-sequence mapping capacity of transformers by relating them to finite transducers, and find that they can express surprisingly large classes of transductions.

Hard Attention POS

Benchmarking Compositionality with Formal Languages

1 code implementation • COLING 2022 • Josef Valvoda, Naomi Saphra, Jonathan Rawski, Adina Williams, Ryan Cotterell

Recombining known primitive concepts into larger novel combinations is a quintessentially human cognitive capability.

Benchmarking Open-Ended Question Answering

Learning with Partially Ordered Representations

no code implementations • WS 2019 • Jane Chandlee, Remi Eyraud, Jeffrey Heinz, Adam Jardine, Jonathan Rawski

This paper examines the characterization and learning of grammars defined with enriched representational models.

Position Relation