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:20240626T180033Z
LOCATION:3004\, 3rd Floor
DTSTART;TZID=America/Los_Angeles:20240626T140000
DTEND;TZID=America/Los_Angeles:20240626T141500
UID:dac_DAC 2024_sess145_RESEARCH651@linklings.com
SUMMARY:LLM-HD: Layout Language Model for Hotspot Detection with GDS Seman
 tic Encoding*
DESCRIPTION:Research Manuscript\n\nyuyang chen, Yiwen Wu, Jingya Wang, Tao
  Wu, Xumin He, Jingyi Yu, and Hao Geng (Shanghai Tech University)\n\nWith 
 the rapid downscaling of technology nodes, \nindustrial flow such as pitch
  reduction, patterning flexibility, and lithography processing variability
  have been challenged. \nLayout hotspot detection is one of the most chall
 enging and critical steps, which requires technology upgrading.\nPattern m
 atching and learning-based detectors are proposed as quick detection metho
 ds. \nHowever, these computer vision (CV) model-based detectors use images
  transformed from layout GDS files as their inputs.  \nIt leads to foregro
 und information (e.g. metal polygons) loss and even distortion when shrink
 ing the image size to fit the model input. \nMoreover, plenty of irrelevan
 t background information such as non-polygon pixels are also fed into the 
 model, \nwhich hinders the fitting of the model and results in a waste of 
 computational resources. \nConcerning the disadvantage of the traditional 
 CV model, we propose a new layout hotspot detection paradigm, \nwhich dire
 ctly detects hotspots on GDS files by exploiting a hierarchical GDS semant
 ic representation scheme and a well-designed pre-trained natural language 
 processing (NLP) model. \nCompared with state-of-the-art works, \nours ach
 ieves better results both on the ICCAD2012 metal layer benchmark and the m
 ore challenging ICCAD2020 via layer benchmark, which demonstrates the effe
 ctiveness and efficiency of our approach.\n\nTopic: Design\n\nKeyword: Des
 ign for Manufacturability and Reliability\n\nSession Chairs: Shao-Yun Fang
  (National Taiwan University of Science and Technology) and Biying Xu (The
  Hong Kong University of Science and Technology (Guangzhou))
END:VEVENT
END:VCALENDAR