BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030655Z
LOCATION:Meeting Room S423+S424\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251218T095400
DTEND;TZID=Asia/Hong_Kong:20251218T100500
UID:siggraphasia_SIGGRAPH Asia 2025_sess147_tog_108@linklings.com
SUMMARY:B4M: Breaking Low-Rank Adapter for Making Content-Style Customizat
 ion
DESCRIPTION:Yu Xu (Chinese Academy of Sciences Institute of Computing Tech
 nology); Fan Tang and Juan Cao (Institute of Computing Technology, Chinese
  Academy of Sciences); Yuxin Zhang (Institute of Automation, Chinese Acade
 my of Sciences); Oliver Deussen (University of Konstanz); weiming Dong (In
 stitute of Automation, Chinese Academy of Sciences); Jintao Li (Institute 
 of Computing Technology, Chinese Academy of Sciences); and Tong-Yee Lee (N
 ational Cheng-Kung University)\n\nThis paper proposes a novel framework fo
 r personalized content-style fusion generation by training content and sty
 le in separated parameter space of low-rank adaptations for pre-trained te
 xt-to-image models. We introduce “partly learnable projection” (PLP) matri
 ces and a “break-for-make” pipeline, achieving superior content-style alig
 nment compared to state-of-the-art methods.\n\nRegistration Category: Full
  Access, Full Access Supporter\n\nSession Chair: Fan Tang (Institute of Co
 mputing Technology, Chinese Academy of Sciences)\n\n
END:VEVENT
END:VCALENDAR