Philip Resnik

About

Philip Resnik is a professor in the Department of Linguistics and a member of the Maryland Language Science Center at the University of Maryland. He also holds affiliate positions in the Department of Computer Science and the Institute for Advanced Computer Studies. His research expertise encompasses computational linguistics, computational modeling, and neurocognitive mechanisms of language processing. Resnik's work involves investigating speech perception, language comprehension, and the neural bases of contextual prediction using methods such as magnetoencephalography (MEG) and functional MRI (fMRI). He has contributed to understanding how local and global contextual models are employed in speech processing, as well as exploring the neural correlates of linguistic and topical prediction. Additionally, Resnik has engaged in evaluating topic models and applying machine learning techniques to predict suicidal ideation and political ideology, demonstrating a broad interest in applying computational methods to linguistic and psychological questions.

Research topics

• Artificial Intelligence
• Machine Learning
• Natural Language Processing
• Computer Science
• Linguistics

Selected publications

• Is Automated Topic Model Evaluation Broken?: The Incoherence of\n Coherence

  arXiv (Cornell University) · 2021 · 67 citations

  Senior authorCorresponding

  Topic model evaluation, like evaluation of other unsupervised methods, can be\ncontentious. However, the field has coalesced around automated estimates of\ntopic coherence, which rely on the frequency of word co-occurrences in a\nreference corpus. Contemporary neural topic models surpass classical ones\naccording to these metrics. At the same time, topic model evaluation suffers\nfrom a validation gap: automated coherence, developed for classical models, has\nnot been validated using human exper…

  DOI

• Large Language Models Are Biased Because They Are Large Language Models

  Computational Linguistics · 2025-01-01 · 19 citations

  articleOpen access1st authorCorresponding

  Abstract This position paper’s primary goal is to provoke thoughtful discussion about the relationship between bias and fundamental properties of large language models (LLMs). I do this by seeking to convince the reader that harmful biases are an inevitable consequence arising from the design of any large language model as LLMs are currently formulated. To the extent that this is true, it suggests that the problem of harmful bias cannot be properly addressed without a serious reconsideration of…

  PublisherDOI

• Neural Dynamics of the Processing of Speech Features: Evidence for a Progression of Features from Acoustic to Sentential Processing

  Journal of Neuroscience · 2025-01-14 · 9 citations

  articleOpen access

  When we listen to speech, our brain's neurophysiological responses "track" its acoustic features, but it is less well understood how these auditory responses are enhanced by linguistic content. Here, we recorded magnetoencephalography responses while subjects of both sexes listened to four types of continuous speechlike passages: speech envelope-modulated noise, English-like nonwords, scrambled words, and a narrative passage. Temporal response function (TRF) analysis provides strong neural evide…

  PublisherOA PDFDOI

• Identification of Long-Term Care Facility Residence From Admission Notes Using Large Language Models

  JAMA Network Open · 2025-05-22 · 4 citations

  articleOpen access

  Importance: An estimated half of all long-term care facility (LTCF) residents are colonized with antimicrobial-resistant organisms, and early identification of these patients on admission to acute care hospitals is a core strategy for preventing intrahospital spread. However, because LTCF exposure is not reliably captured in structured electronic health record data, LTCF-exposed patients routinely go undetected. Large language models (LLMs) offer a promising, but untested, opportunity for extrac…

  PublisherOA PDFDOI

• A Multimodal Framework for the Assessment of the Schizophrenia Spectrum

  2024-09-01 · 3 citations

  article
  PublisherDOI

Recent grants

• RAPID: Advanced Topic Modeling Methods to Analyze Text Responses in COVID-19 Survey Data

  NSF · $192k · 2020–2023

• SoCS: Collaborative Research: Data Driven, Computational Models for Discovery and Analysis of Framing

  NSF · $164k · 2012–2015

Frequent coauthors

• Douglas W. Oard

  32 shared

• Jordan Boyd‐Graber

  25 shared

• Joe Barrow

  21 shared

• Rajiv Jain

  20 shared

• Vlad I. Morariu

  Adobe Systems (United States)

  20 shared

• Varun Manjunatha

  20 shared

• Amber E. Boydstun

  University of California, Davis

  18 shared

• Matthew T. Pietryka

  Florida State University

  17 shared

Labs

• Philip Resnik LabPI

Similar researchers at University of Maryland, College Park

• Colin Phillipsh-49
• Xiaoyu Yangh-46
• Ellen Lauh-39