R
Richard Futrell
Researcher at University of California, Irvine
Publications - 107
Citations - 2607
Richard Futrell is an academic researcher from University of California, Irvine. The author has contributed to research in topics: Computer science & Word order. The author has an hindex of 23, co-authored 88 publications receiving 1801 citations. Previous affiliations of Richard Futrell include Massachusetts Institute of Technology & Apple Inc..
Papers
More filters
Proceedings ArticleDOI
Neural language models as psycholinguistic subjects: Representations of syntactic state.
TL;DR: The authors investigate the extent to which the behavior of neural network language models reflect incremental representations of syntactic state and reveal the specific lexical cues that networks use to update these states, but only the models trained on large datasets are sensitive to subtle lexical clues signaling changes in syntactic states.
Patent
Exemplar-based natural language processing
Richard Futrell,Thomas R. Gruber +1 more
TL;DR: In this paper, an exemplar-based NLP system for NLP is presented, and a semantic edit distance between the first text phrase and the second text phrase in a semantic space can be determined based on one or more of the insertion cost, the deletion cost, and the substitution cost.
Journal ArticleDOI
Lossy‐Context Surprisal: An Information‐Theoretic Model of Memory Effects in Sentence Processing
TL;DR: A new model of incremental sentence processing difficulty that unifies and extends key features of both kinds of models, and demonstrates that dependency locality effects, a signature prediction of memory‐based theories, can be derived from lossy‐context surprisal as a special case of a novel, more general principle called information locality.
Journal ArticleDOI
Composition is the Core Driver of the Language-selective Network
Francis Mollica,Matthew Siegelman,Evgeniia Diachek,Steven Piantadosi,Zachary Mineroff,Richard Futrell,Hope Kean,Peng Qian,Evelina Fedorenko,Evelina Fedorenko,Evelina Fedorenko +10 more
TL;DR: This finding demonstrates that composition is robust to word order violations, and that the language regions respond as strongly as they do to naturalistic linguistic input, providing that composition can take place.
Posted Content
RNNs as psycholinguistic subjects: Syntactic state and grammatical dependency.
TL;DR: It is demonstrated that these models represent and maintain incremental syntactic state, but that they do not always generalize in the same way as humans.