Skip to main content
Skip to main content

Language Science Lunch Talk - Katherine Howitt

Linguistics PhD student Katherine Howitt, standing in front of a chalkboard, smiling at the camera.

Language Science Lunch Talk - Katherine Howitt

Linguistics | Maryland Language Science Center Thursday, February 13, 2025 12:15 pm - 1:30 pm HJ Patterson, 2130

Top Ten Ways to Tell Your LLM is not a Typically Developing Human Child: Understanding large language models contribution to the study of language acquisition

Lunch served at 12:15 PM.

Abstract: This talk will explore parallels (and non-parallels) in the performance of large language models and human language learners.

About: "I am a linguist and a computer scientist. Currently, I am a PhD Student in Linguistics at the University of Maryland where I’m advised by Dr. Jeff Lidz and Dr. Colin Phillips. Broadly, I am interested in language acquisition and psycholinguistics.

I have a number of ongoing projects including looking at four year olds knowledge of strong crossover, school age children’s behavior in the speeded cloze paradigm, 19-month-old’s knowledge of argument vs. adjunct questions, and the usefulness of large language models in language acquisition."

Add to Calendar 02/13/25 12:15:00 02/13/25 13:30:00 America/New_York Language Science Lunch Talk - Katherine Howitt

Top Ten Ways to Tell Your LLM is not a Typically Developing Human Child: Understanding large language models contribution to the study of language acquisition

Lunch served at 12:15 PM.

Abstract: This talk will explore parallels (and non-parallels) in the performance of large language models and human language learners.

About: "I am a linguist and a computer scientist. Currently, I am a PhD Student in Linguistics at the University of Maryland where I’m advised by Dr. Jeff Lidz and Dr. Colin Phillips. Broadly, I am interested in language acquisition and psycholinguistics.

I have a number of ongoing projects including looking at four year olds knowledge of strong crossover, school age children’s behavior in the speeded cloze paradigm, 19-month-old’s knowledge of argument vs. adjunct questions, and the usefulness of large language models in language acquisition."

HJ Patterson false