Bhavuk Jain
Bhavuk Jain is a Tech Lead, Applied AI at Google Android, where he focuses on turning cutting-edge AI research into scalable products used by millions of people worldwide. He also is engaged in active research on LLMs and advanced on-device optimizations, driving innovations in efficient, edge-based computing. He has played a key role in bringing these advanced on-device and cloud-based AI capabilities to Pixel devices and Android through contributions to AOSP, enabling more personalized, intelligent, and context-aware experiences now adopted by global device manufacturers.
Research Areas
Authored Publications
Sort By
Preview abstract
Multimodal large language models (LLMs) integrate and process information from multiple modalities such as text, images, audio, and video, enabling complex tasks such as audio translation and visual question answering. While powerful, this complexity introduces novel vulnerabilities to sophisticated adversarial attacks. This survey paper provides a comprehensive overview of this rapidly expanding field, systematically categorizing attacks that range from manipulations of single modalities (e.g., perturbed images or audio) to those exploiting cross-modal interactions. We overview how these attacks exploit weaknesses in model fusion, attention mechanisms, and representation learning and provided analyses on their potential for real-world consequences.
View details
