Aditya Vaidya

2 accepted papers

2026

Abstraction Induces the Brain Alignment of Language and Speech Models

Emily Cheng, Aditya Vaidya, Richard Antonello

ICML 2026poster

Research has repeatedly demonstrated that intermediate hidden states extracted from large language models and speech audio models predict measured brain response to natural language stimuli. Yet, very little is known about the representation properties that enable this high prediction performance. W…

Cited by 0 Source ↗Scholar ↗

2023

Scaling laws for language encoding models in fMRI

Richard Antonello, Aditya Vaidya, Alexander Huth

NeurIPS 2023poster

Representations from transformer-based unidirectional language models are known to be effective at predicting brain responses to natural language. However, most studies comparing language models to brains have used GPT-2 or similarly sized language models. Here we tested whether larger open-source m…

Cited by 59 Source ↗PDF ↗code ↗Scholar ↗