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:20260728T172800
DTEND;TZID=America/Los_Angeles:20260728T172900
UID:dac_DAC 2026_sess306_WIP3212@linklings.com
SUMMARY:Late Breaking Results: Hardware-Efficient PQ Search Framework for 
 Sparse Coding-Based KV Cache Compression
DESCRIPTION:Sohyeon Kim (Hanyang University); Doohyun Cho (D.Notitia Inc.,
 ); Myounghoon Cho, Donghoon Han, and SeungJae Lee (Dnotitia Inc.); and Ji-
 Hoon Kim (Hanyang University)\n\nThis paper proposes an OPQ-based two-stag
 e search framework for efficient sparse coding in KV cache compression. By
  integrating Optimized Product Quantization (OPQ) with a filter-and-refine
  strategy, our framework identifies a compact candidate subset using compr
 essed metadata before performing exact refinement. Experimental results sh
 ow that the proposed method reduces computational volume by up to 2.46x an
 d memory traffic by 4.5x compared to the Batch-OMP baseline, while maintai
 ning exact support recovery with only 12.5% memory overhead.\n\nTrack: Stu
 dent\n\n
END:VEVENT
END:VCALENDAR