AImpact brand
International Recognition for IIT Postdoctoral Researcher: The HERO Methodology Set as a Reference Point at ICCAD Cadence CAD Contest

International Recognition for IIT Postdoctoral Researcher: The HERO Methodology Set as a Reference Point at ICCAD Cadence CAD Contest

June 16, 2026

The Department of Electrical, Computer Engineering (ECE) of the University of Thessaly announced a major international recognition for Dr. Konstantinos G. Liakos, Postdoctoral Researcher, collaborating faculty member.

The HERO methodology (A novel holistic approach for hardware Trojan detection powered by deep learning), developed by Dr. Liakos, has been featured as a foundational Artificial Intelligence approach in the official problem statement of the 2025 CAD Contest at ICCAD. Organized by Cadence Design Systems, a global leader in electronic design automation (EDA), the prestigious competition attracted a record 247 teams from 12 countries. HERO serves as a key benchmark for Problem A, which focuses on Hardware Trojan Detection at the gate-level netlist.

From a CERN-Backed Project to an Industry Benchmark The HERO methodology began in 2018 under the European ATTRACT initiative (Horizon 2020), under the auspices of CERN. It was fully developed through Dr. Liakos’s doctoral dissertation at the Circuits, Systems Laboratory (CAS lab) of the University of Thessaly, under the supervision of Professor Fotios Plessas , in collaboration with Dr. G. Georgakilas, Dr. S. Moustakidis, Dr. N. Sketopoulos, other CAS lab researchers.

This multi-year research focused on detecting malicious hardware modifications during the pre-silicon stage of ASICs using Deep Learning, Generative Adversarial Networks (GANs). The project also delivered valuable open-source resources to the global scientific community:

GAINESIS: A tool for synthesizing new circuit data. • ATLAS: A tool for classifying infected, healthy circuits. • CAS-HtBase: A database for studying Hardware Trojans.

Global Impact In the official contest text written by Cadence engineers, HERO is cited as a comprehensive detection mechanism leveraging deep neural networks. The evolution of HERO from a CERN-supported PhD project to an international reference point for a premier industry competition highlights the world-class impact of research produced within Greek academic institutions.

Read more: here