BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260624T171340Z
LOCATION:Bldg. 8 - Entrance Hall
DTSTART;TZID=Europe/Stockholm:20260630T173000
DTEND;TZID=Europe/Stockholm:20260630T194500
UID:submissions.pasc-conference.org_PASC26_sess135_posC116@linklings.com
SUMMARY:ACMP12 - Semantic-Aware Implicit Neural Compression for Physics Si
 mulations
DESCRIPTION:Jessica Ezemba (Carnegie Mellon University)\n\nMachine learnin
 g surrogates and data-driven scientific discovery require efficient access
  to simulation data, yet physics simulations generate terabyte-scale datas
 ets. Traditional compression either achieves insufficient ratios or corrup
 ts physics-critical features like conservation laws. Implicit neural repre
 sentations offer a promising alternative, but adoption has been limited by
  lengthy training times and dataset-specific fitting. We present SINCPS, l
 everaging wafer-scale computing to train models in 2 to 3 hours each. Acro
 ss 22 datasets from The Well benchmark, we achieve 150× to 25,000× compres
 sion. Turbulent flows and 3D data remain challenging (13 dB), but half of 
 the datasets exceed 20 dB, enabling integration of large simulation archiv
 es into discovery workflows.\n\n
END:VEVENT
END:VCALENDAR