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:20260805T205522Z
LOCATION:Room 408 B
DTSTART;TZID=America/Los_Angeles:20260720T090000
DTEND;TZID=America/Los_Angeles:20260720T100500
UID:siggraph_SIGGRAPH 2026_sess127@linklings.com
SUMMARY:Numerical Geometry a la Mode
DESCRIPTION:Boundary-aware Neural Model Reduction for PDEs\n\nNeural eigen
 analysis enables reduced-order modeling across shape families but is limit
 ed to Neumann boundaries. We extend it to Dirichlet, Robin, and mixed cond
 itions by treating boundary configurations as inputs, forming a unified sh
 ape–boundary space for consistent spectral analysis and simu...\n\n\nLi Li
 ao, Pengfei Shen, and Yifan Peng (The University of Hong Kong)\n----------
 -----------\nLearning Laplacian Eigenspace with Mass-Aware Neural Operator
 s on Point Clouds\n\nNEO is a neural framework for fast Laplace-Beltrami s
 pectral analysis on 3D point clouds. Rather than computing eigenvectors wi
 th expensive iterative solvers, it predicts low-frequency eigenspaces dire
 ctly from geometry, enabling near-linear scaling, robustness to irregular 
 sampling, and accurate ze...\n\n\nZherui Yang (University of Science and T
 echnology of China); Tao Du (Tsinghua University, Shanghai Qi Zhi Institut
 e); and Ligang Liu (University of Science and Technology of China)\n------
 ---------------\nEfficient Multiscale Lanczos Eigenpair Extraction\n\nWe e
 xtend the implicitly restarted Lanczos method to a multiscale context usin
 g arbitrary multigrids (algebraic, geometric) and we demonstrate the gain 
 in performance and robustness on a variety of application scenarios.\n\n\n
 Theo Braune (Centre National de la Recherche Scientifique - Laboratoire d'
 informatique de l'École Polytechnique (LIX), Adobe Research) and Jérémie D
 umas and Jean-Marc Thiery (Adobe)\n---------------------\nComponent Modes 
 Synthesis Method with Multiple Partitions for Large-scale Eigenvalue Probl
 em\n\nTo solve ultra-large-scale modal analysis problems, we extend Compon
 ent Mode Synthesis (CMS) methods to use multiple partitions to generate mo
 re efficient local eigenbasis functions and reduce the computational cost 
 of the low-frequency interface modes. We obtained the lowest 1500 eigenmod
 es of a mo...\n\n\nChongyao Zhao (State Key Lab of CAD and CG, Zhejiang Un
 iversity); Junzhou Yin (Independent Researcher); and Hujun Bao and Jin Hua
 ng (State Key Lab of CAD and CG, Zhejiang University)\n-------------------
 --\nNumerical Geometry a la Mode - Interactive Discussion\n\nAfter the sum
 mary presentations, attendees will participate in an interactive discussio
 n. Outside the room will be a series of poster boards for authors to gathe
 r around with the audience. Authors are invited to bring any material rela
 ted to their paper that could instigate further conversation such...\n\n\n
 Interest Area: Research & Education\n\nKeyword: Artificial Intelligence/Ma
 chine Learning, Geometry, Modeling, Simulation\n\nRegistration Category: F
 ull Conference Supporter, Full Conference\n\nSession Chair: Yiying Tong (M
 ichigan State University)
END:VEVENT
END:VCALENDAR