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:20260730T152728Z
LOCATION:Mtg Room 202C
DTSTART;TZID=America/Los_Angeles:20260728T115000
DTEND;TZID=America/Los_Angeles:20260728T120300
UID:dac_DAC 2026_sess158_RESEARCH641@linklings.com
SUMMARY:TorR: Towards Brain-Inspired Task-Oriented Reasoning via Cache-Ori
 ented Algorithm-Architecture Co-design
DESCRIPTION:Hyunwoo Oh (University of California, Irvine); SungHeon Jeong 
 (UCI); Suyeon Jang, Hanning Chen, Sanggeon Yun, and Tamoghno Das (Universi
 ty of California, Irvine); and Mohsen Imani (University of California Irvi
 ne)\n\nTask-oriented object detection on CLIP enables open-vocabulary, pro
 mpt-driven semantics but dense alignment and memory traffic block real-tim
 e edge use. We propose TorR, a brain-inspired algorithm–architecture co-de
 sign that replaces CLIP-style dense matching with hyperdimensional associa
 tive reasoning and exploits temporal coherence. TorR combines HDC similari
 ty, graph reasoning with query caching and delta updates, and a lane-scala
 ble, precision-gated item memory with RT-30/RT-60 control. A 28 nm TorR ac
 celerator delivers real-time detection with millijoule energy and competit
 ive AP at orders-of-magnitude lower energy.\n\nTopics: Systems\n\nTrack: S
 YS2. Design of Cyber-Physical Systems and IoT\n\nSession Chairs: Hokeun Ki
 m (Arizona State University) and BaekGyu Kim (Daegu Gyeongbuk Institute of
  Science and Technology)\n\n
END:VEVENT
END:VCALENDAR
