activity
20242026
collaborators

9 papers

cs.LG2026

Coupled Query-Key Dynamics for Attention

Barak Gahtan, Alex M. Bronstein

Standard scaled dot-product attention computes scores from static, independent projections of the input. We show that evolving queries and keys \emph{jointly} through shared learne…

cs.LG2026

Effective Sample Size and Generalization Bounds for Temporal Networks

Barak Gahtan, Alex M. Bronstein

Learning from time series is fundamentally different from learning from i.i.d.\ data: temporal dependence can make long sequences effectively information-poor, yet standard evaluat…

cs.CE2026

Wildfire Simulation with Differentiable Randers-Finsler Eikonal Solvers

Barak Gahtan, Jacob Shpund, Alex M. Bronstein

Fast and differentiable solvers for anisotropic and asymmetric distance fields are a key primitive in geometry processing, enabling gradient-based optimization over metrics, drift…

cs.NI2025

Exploring QUIC Dynamics: A Large-Scale Dataset for Encrypted Traffic Analysis

Barak Gahtan, Robert J. Shahla, Alex M. Bronstein +1

The increasing adoption of the QUIC transport protocol has transformed encrypted web traffic, necessitating new methodologies for network analysis. However, existing datasets lack…

cs.LG2025

From Lab to Wrist: Bridging Metabolic Monitoring and Consumer Wearables for Heart Rate and Oxygen Consumption Modeling

Barak Gahtan, Sanketh Vedula, Gil Samuelly Leichtag +2

Understanding physiological responses during running is critical for performance optimization, tailored training prescriptions, and athlete health management. We introduce a compre…

cs.LG2025

Estimating the Number of HTTP/3 Responses in QUIC Using Deep Learning

Barak Gahtan, Robert J. Shahla, Reuven Cohen +1

QUIC, a new and increasingly used transport protocol, enhances TCP by offering improved security, performance, and stream multiplexing. These features, however, also impose challen…