The iCAS Lab Ph.D. Students, Mahsa Ardakani and Jinendra Malekar, have received the Best Paper Award at the ACM GLSVLSI 2026 Conference for their paper titled "PiCASO: A Real-time Pi-Integrated Conversational AI System with Optimized LLMs."

In this paper, which is part of a project funded by the NSF CAREER program, Researchers from the iCAS Lab have demonstrated a new approach for bringing powerful conversational AI to low-cost edge devices through advanced quantization techniques. Their work enabled PiCASo to deliver real-time, fully offline conversational AI on a Raspberry Pi 5, eliminating the need for cloud connectivity while maintaining strong performance. This achievement opens the door to practical, privacy-preserving AI applications in social robotics, education, accessibility, and other embedded systems where computing resources are limited.