About

Tim Hunter is a Professor in the Department of Linguistics at the University of California, Los Angeles. His research focuses on formal and computational approaches to syntax and semantics, including the investigation of grammatical hypotheses through integration with language processing and acquisition models. Hunter has contributed to understanding the generative capacity of linguistic formalisms, the nature of syntactic variation, and the role of phonological and syntactic factors in sentence probability. His work also explores the semantics of quantificational words like 'most' and their cognitive underpinnings, as well as the properties of ellipsis and argument/adjunct distinctions in syntax. Hunter has developed formal methods in experimental syntax, examining the effects of derivational operations, and has engaged in theoretical work on the syntax-semantics interface, including the effects of remnant movement, adjuncts, and the argument/adjunct distinction. His research extends to the study of noise-tolerant language learning models, the interaction of phonological and syntactic constraints, and the integration of grammatical hypotheses into broader cognitive models. Hunter has also contributed to the understanding of linguistic meaning as cognitive instructions and has conducted experiments on language acquisition and the learnability of determiners. He has taught courses at various levels, including computational linguistics, syntax, and semantics, and has presented his work at numerous conferences and workshops.

Research topics

• Computer Science
• History
• Artificial Intelligence
• Natural Language Processing
• Linguistics
• Psychology
• Mathematics
• Geography
• Cognitive science
• Cognitive psychology

Selected publications

• Computational Approaches to Syntactic Variation

  Cambridge University Press eBooks · 2025-09-04

  book-chapter1st authorCorresponding
  PublisherDOI

• Modeling regularization in language acquisition as noise-tolerant grammar selection

  Cognition · 2025-11-13 · 1 citations

  articleSenior author
  PublisherDOI

• Kallini et al. (2024) Do Not Compare Impossible Languages with Constituency-based Ones

  Computational Linguistics · 2025-01-01

  articleOpen access1st authorCorresponding

  Abstract A central goal of linguistic theory is to find a precise characterization of the notion “possible human language”, in the form of a computational device that is capable of describing all and only the languages that can be acquired by a typically developing human child. The success of recent large language models (LLMs) in NLP applications arguably raises the possibility that LLMs might be computational devices that meet this goal. This would only be the case if, in addition to succeeding in learning human languages, LLMs struggle to learn “impossible” human languages. Kallini et al. (2024) conducted experiments aiming to test this by training GPT-2 on a variety of synthetic languages, and found that it learns some more successfully than others. They present these asymmetries as support for the idea that LLMs’ inductive biases align with what is regarded as “possible” for human languages, but the most significant comparison has a confound that makes this conclusion unwarranted.

  PublisherDOI

• BARNARD’S MYSTERIOUS STAR NEAR VENUS: A STRANGE INTERLOPER NOTED DURING A SATELLITE SEARCH

  Journal of Astronomical History and Heritage · 2025-09-01

  article

  Edward Emerson Barnard is regarded as one of the greatest astronomical observers of all time, who discovered numerous comets, the fifth satellite of Jupiter, and pioneered the wide-angle photography of the Milky Way. Given his well-earned reputation as the most judicious and reliable observer of his time, an observation of a seventh-magnitude star near Venus with the 36-inch refractor of Lick Observatory on 13 August 1892, which has never been explained, has long intrigued and baffled astronomers and historians of astronomy. Barnard recorded it incidentally while searching for Venusian satellites, and though at the time no doubt regarding it as an ordinary field star, he later established it did not correspond in position to any known star. As was usual with him for doubtful observations, he did not publish, at first. Yet later he did, for reasons unknown, fourteen years later. In this paper, we consider Barnard’s observation in the context of his career and struggle to establish his reputation as an observer, and also summarize the results of our own investigations to try to determine, at long last, just what he likely saw. In particular, we have exhaustively listed and tested all possible astrophysical, atmospheric, and telescopic mechanisms for a transient source (including novae, comets, asteroids, variable stars of all types, and various classes of ghost images), and we refute them all at a high confidence level. The only remaining possibility is that the problem comes from some ordinary error in Barnard’s report. We conclude that Barnard must have seen the 11th magnitude field star (TYC 1348-1601-1) and simply reported its magnitude as brighter than it actually is (due to his lack of any comparison star in twilight and his unfamiliarity with the 36-inch telescope).

  PublisherDOI

• MG1-1995959: An Eclipsing Binary Star with a Pronounced O’Connell Effect

  The Astrophysical Journal · 2025-06-17 · 1 citations

  articleOpen accessCorresponding

  Abstract MG1-1995959 is a short-period (∼0.586 day) eclipsing binary, exhibiting at times an extraordinarily large O’Connell effect (OE), for which this paper reports its discovery, time series photometry, and spectroscopic observation. The photometry is scattered across years 2001–2023, with spectroscopic observations from 2023. Photometric curves exhibit classical OE asymmetries, whereby maxima following primary and secondary eclipses are of different magnitudes. Short time span photometric time series provide static characterizations of OE stars yielding incomplete representations of critical macrobehavior. In contrast, MG1-1995959 photometry is of sufficient duration to indicate pronounced changes in such behavior, suggesting that intense observing programs can provide key benchmark data to better inform complex modeling efforts. Spectra indicate the primary to be K1 Vk(e), and the secondary type is ∼K5 V. PHOEBE modeling of the light curve from 2014 enables estimation of key stellar parameters (primary/secondary): T eff [K] (5110/4260), and, in solar units, mass (0.86/0.68) and radius (0.93/0.65). The inclination of the system was found to be ∼78°. PHOEBE modeling of light curves from 2013, 2022, and 2023 is presented, demonstrating (1) the light curves can be well fit by invoking distributions of star spots, (2) optimal spot sizes and distributions vary dramatically as a function of time, and (3) this is a powerful tool with which to investigate the short term evolution of this star. MG1-1995959 is an accessible star with dynamic and unusual behavior deserving of intensive study to better understand nuances of the OE and the nature of late-type main-sequence stars.

  PublisherDOI

• Kallini et al. (2024) do not compare impossible languages with constituency-based ones

  arXiv (Cornell University) · 2024-10-16

  preprintOpen access1st authorCorresponding

  A central goal of linguistic theory is to find a precise characterization of the notion "possible human language", in the form of a computational device that is capable of describing all and only the languages that can be acquired by a typically developing human child. The success of recent large language models (LLMs) in NLP applications arguably raises the possibility that LLMs might be computational devices that meet this goal. This would only be the case if, in addition to succeeding in learning human languages, LLMs struggle to learn "impossible" human languages. Kallini et al. (2024; "Mission: Impossible Language Models", Proc. ACL) conducted experiments aiming to test this by training GPT-2 on a variety of synthetic languages, and found that it learns some more successfully than others. They present these asymmetries as support for the idea that LLMs' inductive biases align with what is regarded as "possible" for human languages, but the most significant comparison has a confound that makes this conclusion unwarranted. In this paper I explain the confound and suggest some ways forward towards constructing a comparison that appropriately tests the underlying issue.

  PublisherOA PDFDOI

• Formal Methods in Experimental Syntax

  Oxford University Press eBooks · 2023-09-18

  book-chapter1st authorCorresponding

  Abstract This chapter provides an overview of some ways to bring evidence from experimental language-processing work to bear on theories of syntax, where the relevant syntactic theories are formulated as explicit, self-contained formal grammars. The focus is on linking hypotheses that take the form of complexity metrics, at differing levels of abstraction. At one level are information-theoretic metrics such as surprisal and entropy reduction, where a grammar’s role is to define a probability distribution over generated expressions; a key point emphasized here is the way a grammar’s generative mechanisms constrain the shapes of these distributions. At a lower, explicitly algorithmic, level of abstraction are metrics based on the memory load incurred by particular parsing algorithms. The concluding section outlines applications of these ideas to modern syntactic theory.

  PublisherDOI

• Astronomical Catalogs: An Overview

  Patrick Moore's practical astronomy series/˜The œPatrick Moore practical astronomy series · 2023-01-01

  book-chapter1st authorCorresponding
  PublisherDOI

• The Future of Experimental Syntax

  Oxford University Press eBooks · 2023-09-18

  book-chapter

  Abstract In this concluding chapter of the handbook, each contributor has written a 500 word mini-essay presenting their view of the future of experimental syntax, from the important theoretical questions on the horizon, to the methodological challenges that experimental syntacticians need to solve to answer those questions. The hope is that this final chapter will serve both as concrete inspiration for future studies in experimental syntax and as a benchmark for measuring the success of the field in the years to come.

  PublisherDOI

• On regular copying languages

  Journal of Language Modelling · 2023-07-21 · 1 citations

  articleOpen accessSenior author

  This paper proposes a formal model of regular languages enriched with unbounded copying. We augment finite-state machinery with the ability to recognize copied strings by adding an unbounded memory buffer with a restricted form of first-in-first-out storage. The newly introduced computational device, finite-state buffered machines (FS-BMs), characterizes the class of regular languages and languages de-rived from them through a primitive copying operation. We name this language class regular copying languages (RCLs). We prove a pumping lemma and examine the closure properties of this language class. As suggested by previous literature (Gazdar and Pullum 1985, p.278), regular copying languages should approach the correct characteriza-tion of natural language word sets.

  PublisherOA PDFDOI

Frequent coauthors

• Mihra S. Taljanovic

  University of Arizona

  49 shared

• Lee F. Rogers

  21 shared

• Joseph E. Sheppard

  19 shared

• Marci D. Jones

  University of Massachusetts Chan Medical School

  18 shared

• Daniel N. Switlick

  University of Arizona

  16 shared

• Jeffrey Lidz

  University of Maryland, College Park

  15 shared

• Justin Halberda

  Johns Hopkins University

  12 shared

• Paul M. Pietroski

  Rutgers Sexual and Reproductive Health and Rights

  12 shared

Education

• Ph.D., Syntax and Semantics

  University of Maryland

  2010

• M.A., Linguistics

  University of Maryland

• B.A., Linguistics

  University of Maryland