What Do VLMs NOTICE? A Mechanistic Interpretability Pipeline for Gaussian-Noise-free Text-Image Corruption and Evaluation

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Golovanevsky, Michal, Rudman, William, Palit, Vedant, Singh, Ritambhara, Eickhoff, Carsten
Format:	Preprint
Published:	2024
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2406.16320