A fusion of vision DNN and LLM representations predicts human EEG responses to images better than either model alone, with distinct temporal and spectral signatures.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
q-bio.NC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The time course of visuo-semantic representations in the human brain is captured by combining vision and language models
A fusion of vision DNN and LLM representations predicts human EEG responses to images better than either model alone, with distinct temporal and spectral signatures.