Boris Katz

A Pilot Study Exploring the Alignment of Humans and CNN During Perception of Social Interactions

Guido Vallarino, Lucia Schiatti, Matteo Moro, Yen-Ling Kuo et autres

it, us, sg (code pays fourni par la source)

Investigating the Impact of Cerebral vs Peripheral Visual Issues on Emotion Recognition Using a New Silicone-based Face Mask Prototype

Lucia Schiatti, Cristiano Cuppini, Federica Morelli, Giulia Cappagli et autres

This paper introduces an innovative silicone-based semi-transparent face mask prototype, to cope with limitations of the standard face masks massively adopted as individual protection devices (IPDs) during the COVID-19 pandemic. In particular, the prototype aimed at complying with health-related requirements while preserving …

it, us (code pays fourni par la source)

Population Transformer: Learning Population-level Representations of Neural Activity.

Geeling Chau, Christopher Wang, Sabera Talukder, Vighnesh Subramaniam et autres

We present a self-supervised framework that learns population-level codes for arbitrary ensembles of neural recordings at scale. We address key challenges in scaling models with neural time-series data, namely, sparse and variable electrode distribution across subjects and datasets. The Population Transformer (PopT) …

us (code pays fourni par la source)

Training the Untrainable: Introducing Inductive Bias via Representational Alignment

Vighnesh Subramaniam, David Mayo, Colin Conwell, Tomaso Poggio et autres

We demonstrate that architectures which traditionally are considered to be ill-suited for a task can be trained using inductive biases from another architecture. We call a network untrainable when it overfits, underfits, or converges to poor results even when tuning their hyperparameters. …

us, ru (code pays fourni par la source)

Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli

Christopher Wang, Adam Yaari, Vighnesh Subramaniam, Dana Rosenfarb et autres

We present the Brain Treebank, a large-scale dataset of electrophysiological neural responses, recorded from intracranial probes while 10 subjects watched one or more Hollywood movies. Subjects watched on average 2.6 Hollywood movies, for an average viewing time of 4.3 hours, and a …

BrainBits: How Much of the Brain are Generative Reconstruction Methods Using?

David Mayo, Christopher Wang, Asa Harbin, Abdulrahman Alabdulkareem et autres

When evaluating stimuli reconstruction results it is tempting to assume that higher fidelity text and image generation is due to an improved understanding of the brain or more powerful signal extraction from neural recordings. However, in practice, new reconstruction methods could improve …

Revealing Vision-Language Integration in the Brain with Multimodal Networks

Vighnesh Subramaniam, Colin Conwell, Christopher Wang, Gabriel Kreiman et autres

We use (multi)modal deep neural networks (DNNs) to probe for sites of multimodal integration in the human brain by predicting stereoen-cephalography (SEEG) recordings taken while human subjects watched movies. We operationalize sites of multimodal integration as regions where a multimodal vision-language model …

br, us (code pays fourni par la source)

SecureLLM: Using Compositionality to Build Provably Secure Language Models for Private, Sensitive, and Secret Data

Christian M Arnold, Yerim Lee, Pieter M Feenstra, Boris Katz et autres

Traditional security mechanisms isolate resources from users who should not access them. We reflect the compositional nature of such security mechanisms back into the structure of LLMs to build a provably secure LLM; that we term SecureLLM. Other approaches to LLM safety …

Brain Treebank: Large-scale intracranial recordings from naturalistic language stimuli

Christopher Wang, Adam Yaari, Aaditya Singh, Vighnesh Subramaniam et autres

BrainBits: How Much of the Brain are Generative Reconstruction Methods Using?

David Mayo, Christopher Wang, Asa Harbin, Abdulrahman Alabdulkareem et autres

Modeling Visual Impairments with Artificial Neural Networks: a Review

Lucia Schiatti, Monica Gori, Martin Schrimpf, Giulia Cappagli et autres

We present an approach to bridge the gap between the computational models of human vision and the clinical practice on visual impairments (VI). In a nutshell, we propose to connect advances in neuroscience and machine learning to study the impact of VI …

it, us, ch (code pays fourni par la source)

Zero-Shot Linear Combinations of Grounded Social Interactions with Linear Social MDPs

Ravi Tejwani, Yen‐Ling Kuo, Tianmin Shu, Bennett Stankovits et autres

Humans and animals engage in rich social interactions. It is often theorized that a relatively small number of basic social interactions give rise to the full range of behavior observed. But no computational theory explaining how social interactions combine together has been …

ru, us (code pays fourni par la source)