BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260730T152637Z
LOCATION:
DTSTART;TZID=America/Los_Angeles:20260728T123800
DTEND;TZID=America/Los_Angeles:20260728T123800
UID:dac_DAC 2026_sess315_RESEARCH2536@linklings.com
SUMMARY:NetDeTox: Adversarial and Efficient Evasion of Hardware-Security G
 NNs via RL-LLM Orchestration
DESCRIPTION:Zeng Wang and Minghao Shao (New York University); Akashdeep Sa
 ha (New York University, Abu Dhabi); Ramesh Karri (NYU); and Johann Knecht
 el, Muhammad Shafique, and Ozgur Sinanoglu (New York University, Abu Dhabi
 )\n\nGraph neural networks (GNNs) are widely used in hardware security but
  vulnerable to adversarial netlist rewrites. Existing adversarial approach
 es incur high overheads. We present NetDeTox, an automated framework combi
 ning large language models (LLMs) with reinforcement learning (RL) for eff
 icient rewriting. The RL agent identifies GNN-critical components while th
 e LLM devises functionality-preserving rewrites that diversify motifs. Ite
 rative feedback minimizes overheads. Compared to state-of-the-art AttackGN
 N, NetDeTox degrades all security schemes with fewer rewrites and substant
 ially lower area overheads (54.50%, 25.44%, and 41.04% reductions for GNN-
 RE, GNN4IP, and OMLA). For larger circuits, NetDeTox even reduces original
  area, demonstrating practical scalability.\n\nTopics: Quantum\n\nTrack: D
 ES6. Quantum Computing, Systems\n\nSession Chairs: Hokeun Kim (Arizona Sta
 te University); BaekGyu Kim (Daegu Gyeongbuk Institute of Science and Tech
 nology); Xiang Chen (Peking University); Sangyeob Kim (Yonsei University);
  Robert M. Radway (University of Pennsylvania); Prashanth Mohan (Universit
 y of Pittsburgh); Xiaoxuan Yang (University of Virginia, Stanford Universi
 ty); Ramtin Zand (University of South Carolina); Hadi Kamali (University o
 f Central Florida); Jiafeng Xie (Villanova University); Prabuddha Chakrabo
 rty (University of Maine); Junichiro Kadomoto (University of Tokyo); and Y
 osuke Ueno (RIKEN)\n\n
END:VEVENT
END:VCALENDAR