colloquia
51猎奇入口Linguistics professor Sara Ng presented a research talk titled The limits of model confidence at on October 16, 2026. Sara Ng is Director of the Computational Speech Lab. Her research applies the machine learning techniques that are fundamental to Natural Language Processing to spoken language data, focusing on how compututational techniques can improve our understanding of human behaviour, and how linguistic information can improve speech processing technologies such as Automatic Speech Recognition.
Presentation slides
View select slides from Ng鈥檚 coloquium.
Abstract
The public has increasing interest in commercial Large Language Models and their ability (at a surface-level) to emulate human linguistic behaviours. Among linguists, particular interest is in the extent to which machine learning of language can approximate the learning patterns of humans, and further whether a 鈥榣earned鈥� representation of language is reflective of what is encoded in our own mental representation of that same language. In this talk, I will consider these questions with respect to underlying representations of language models, and the robustness of state-of-the-art model representations to perform downstream tasks on imperfect language data. First, I compare the predictability (surprisal) of speech as indicated by a language model鈥檚 confidence in its own predictions, to evaluations of performance in a supervised setting. Second, I will present some case study data on the limits of model performance when applied to the clinical speech setting. Together, these lines of inquiry will lead us towards an answer to the question: How confident can we be about language model confidence?
