Arun Balaji Buduru

The paper introduces RoboKA, a KAN-based multimodal fusion framework, and a synthetic dataset (Robo-SAr) to significantly improve robocall surveillance by modeling complex adversarial interactions between acoustic and linguistic cues.

The paper introduces DECKER, a domain-invariant framework that significantly improves cross-keyboard keystroke inference by normalizing device variations and leveraging linguistic context, demonstrating the persistent security risk of acoustic side-channel attacks.