← Search

Mayukh Deb

2 accepted papers

2025

TopoNets: High performing vision and language models with brain-like topography

ICLR 2025spotlight

Neurons in the brain are organized such that nearby cells tend to share similar functions. AI models lack this organization, and past efforts to introduce topography have often led to trade-offs between topography and task performance. In this work, we present *TopoLoss*, a new loss function that pr…

2023

ATMAN: Understanding Transformer Predictions Through Memory Efficient Attention Manipulation

NeurIPS 2023poster

Generative transformer models have become increasingly complex, with large numbers of parameters and the ability to process multiple input modalities. Current methods for explaining their predictions are resource-intensive. Most crucially, they require prohibitively large amounts of additional memor…