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:Exhibit Hall
DTSTART;TZID=America/Los_Angeles:20260728T172000
DTEND;TZID=America/Los_Angeles:20260728T172000
UID:dac_DAC 2026_sess306_WIP3215@linklings.com
SUMMARY:Late Breaking Results: A Neuromorphic Accelerator for Efficient Mu
 lti-Head Attention Processing in Spiking Vision Transformers
DESCRIPTION:Rachmad Vidya Wicaksana Putra (New York University (NYU) Abu D
 habi), Amirhesam Jafari Rad (University of Tehran), and Muhammad Shafique 
 (New York University Abu Dhabi (NYUAD))\n\nSpiking Vision Transformers (SV
 iTs) are developed as an energy-efficient alternative to conventional ViTs
 . To maximize efficiency gains of SViT processing, we propose MorphAtt, a 
 novel digital SViT accelerator that expedites the inference through stream
 lined processing. Specifically, it processes multi-head self-attention (MH
 SA) operations using spiking Query-Key-Value, spiking self-attention, and 
 reparameterization modules, as well as inter-module buffers to mitigate tr
 affic congestion in on-chip memory accesses. Experimental results show tha
 t, MorphAtt achieves 792-1605 GOPS of throughput, while incurring ∼39-55 m
 W of power consumption and 1.5 mm2 of area, which lead to 20.3-29.1 TOPS/W
  of energy efficiency. These results demonstrate that, our MorphAtt offers
  better performance and efficiency trade-offs than state-of-the-art, thus 
 enabling highly energy-efficient vision-based systems at the edge.\n\nTrac
 k: Student\n\n
END:VEVENT
END:VCALENDAR