3 papers
cs.LG2026
Mechanistic Interpretability Reveals Shared Causal Subspaces in Brain-to-Speech Decoders
Maryam Maghsoudi, Ayushi Mishra, Sanghamitra Dutta
Decoding covert speech, such as mimed or imagined, from brain activity is harder than decoding vocalized speech. Cross-modal transfer, where information from one speech form helps…
cs.LG2026
Mechanistic Interpretability of Brain-to-Speech Models Across Speech Modes
Maryam Maghsoudi, Ayushi Mishra
Brain-to-speech decoding models demonstrate robust performance in vocalized, mimed, and imagined speech; yet, the fundamental mechanisms via which these models capture and transmit…
cs.SD2025
Spatial Audio Processing with Large Language Model on Wearable Devices
Ayushi Mishra, Yang Bai, Priyadarshan Narayanasamy +2
Integrating spatial context into large language models (LLMs) has the potential to revolutionize human-computer interaction, particularly in wearable devices. In this work, we pres…