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:20260402T024533Z
LOCATION:3006\, Level 3
DTSTART;TZID=America/Los_Angeles:20250623T110000
DTEND;TZID=America/Los_Angeles:20250623T111500
UID:dac_DAC 2025_sess139_RESEARCH712@linklings.com
SUMMARY:INSIGHT: A Universal Neural Simulator Framework for Analog Circuit
 s with Autoregressive Transformers
DESCRIPTION:Souradip Poddar (The University of Texas at Austin), Youngmin 
 Oh (Samsung), Yao Lai and Hanqing Zhu (The University of Texas at Austin),
  Bosun Hwang (Samsung), and David Z. Pan (The University of Texas at Austi
 n)\n\nThis paper introduces INSIGHT, a data-efficient, adaptive, high-fide
 lity, technology-agnostic universal neural simulator framework that formul
 ates analog performance prediction as an autoregressive sequence generatio
 n task to accurately predict performance across diverse circuits. INSIGHT 
 achieves test R2 scores ≥0.95, outperforming existing neural surrogates. C
 ross-technology transfer learning experiments show that INSIGHT can preser
 ve model performance with ~60% less training data. Low-Rank Adaptation (Lo
 RA) integration further reduces memory footprint by ~42% and training time
  by ~25%, maintaining high performance. Our experiments show that INSIGHT-
 based RL sizing framework achieves ~100-1000X lower simulation costs over 
 existing sizing methods for identical benchmarks and target specifications
 .\n\nTopics: EDA\n\nTracks: EDA6: Analog CAD, Simulation, Verification and
  Test\n\nSession Chairs: Arindam Basu (School of Computer Science and Engi
 neering, Nanyang Technological University) and Markus Olbrich (Leibniz Uni
 versity Hannover)\n\n
END:VEVENT
END:VCALENDAR