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:20260730T152640Z
LOCATION:
DTSTART;TZID=America/Los_Angeles:20260728T175300
DTEND;TZID=America/Los_Angeles:20260728T175300
UID:dac_DAC 2026_sess317_RESEARCH1628@linklings.com
SUMMARY:AttackonCTF: Defending Hardware Security Competition Benchmarks in
  the Age of LLMs
DESCRIPTION:Mohamadreza Rostami (Technical University of Darmstadt), Nikhi
 lesh Singh (Technische Universität Darmstadt), Stephen Muttathil (Texas A&
 M University), Lichao Wu (Technical University of Darmstadt), Chen Chen (T
 exas A&M University), Huimin Li (Technical University of Darmstadt), Jeyav
 ijayan Rajendran (Texas A&M University), and Ahmad-Reza Sadeghi (Technisch
 e Universitaet Darmstadt)\n\nHardware security competitions such as HackTh
 eSilicon serve as benchmarking platforms for evaluating vulnerability dete
 ction methods and training human and AI. However, our study reveals that L
 LMs threaten their validity. Instead of genuine security reasoning, detect
 ors exploit a diff-style syntactic comparison, achieving an 83% detection 
 rate, undermining fair evaluation. To mitigate this, we propose the first 
 LLM-oriented, semantics-preserving obfuscation framework for these benchma
 rks. Unlike IP-protection approaches, it applies human-readable transforma
 tions and controlled diff-noise while preserving functionality. On HackThe
 Silicon, the framework reduces LLM-based detection accuracy by 50% with on
 ly 10% obfuscation and by 78.6% under complete obfuscation, restoring benc
 hmark reliability.\n\nTrack: EDA7-II. Physical Design and Verification, De
 sign\n\nSession Chairs: Christian Pilato (Politecnico di Milano); Mohsen I
 mani (University of California, Irvine); Flavio Ponzina (San Diego State U
 niversity); Amin Farshidi (Cadence Design Systems, Inc.); Yonggan Fu (Geor
 gia Institute of Technology, NVIDIA); Dimitrios Tsaras (Huawei, Hong Kong 
 University of Science and Technology (HKUST)); Anh Nguyen (Fujitsu Limited
 ); Ruiyang Qin (Villanova University); Jiliang Zhang (Hunan University); M
 eng Li (Peking University); and Rassul Bairamkulov (Advanced Micro Devices
  (AMD))\n\n
END:VEVENT
END:VCALENDAR