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:20260728T174400
DTEND;TZID=America/Los_Angeles:20260728T174500
UID:dac_DAC 2026_sess317_RESEARCH012@linklings.com
SUMMARY:ModuPlace: LLM-Assisted Modular PCB Placement via Preference-Optim
 ized Constraint Graph Generation
DESCRIPTION:Yaohui Han (The Chinese University of Hong Kong); Beichen Li (
 The Chinese University of Hong Kong, Shenzhen); Mingyang Zhao (Central Sou
 th University); Rongliang Fu (The Chinese University of Hong Kong); Qunson
 g Ye (Shenzhen index tech.); Tinghuan Chen (The Chinese University of Hong
  Kong, Shenzhen); and Bei Yu and Tsung-Yi Ho (The Chinese University of Ho
 ng Kong)\n\nDespite the existence of various automated PCB placement frame
 works, the industry still relies heavily on human engineers. This is becau
 se these frameworks cannot perform placement according to specific user re
 quirements and preferences, which limits their flexibility and industrial 
 adoption. Moreover, the existing frameworks lack a modular perspective in 
 placement, resulting in outputs that often perform poorly and fail to meet
  practical requirements. To address these problems, we propose ModuPlace, 
 leveraging the powerful capabilities of LLMs to perform modular PCB placem
 ent with preference-optimized constraint graph generation. In ModuPlace, t
 he entire component set is partitioned into modules at different granulari
 ties, enabling hierarchical and modular placement. Moreover, ModuPlace uti
 lizes fine-tuning and preference optimization to enhance the quality of co
 nstraint graph generation and align the results with those of human expert
 s. Experimental results demonstrate that ModuPlace outperforms all baselin
 es, achieving superior placement quality across all metrics.\n\nTrack: EDA
 7-II. Physical Design and Verification, Design\n\nSession Chairs: Christia
 n Pilato (Politecnico di Milano); Mohsen Imani (University of California, 
 Irvine); Flavio Ponzina (San Diego State University); Amin Farshidi (Caden
 ce Design Systems, Inc.); Yonggan Fu (Georgia Institute of Technology, NVI
 DIA); Dimitrios Tsaras (Huawei, Hong Kong University of Science and Techno
 logy (HKUST)); Anh Nguyen (Fujitsu Limited); Ruiyang Qin (Villanova Univer
 sity); Jiliang Zhang (Hunan University); Meng Li (Peking University); and 
 Rassul Bairamkulov (Advanced Micro Devices (AMD))\n\n
END:VEVENT
END:VCALENDAR