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:20260730T152638Z
LOCATION:
DTSTART;TZID=America/Los_Angeles:20260728T125900
DTEND;TZID=America/Los_Angeles:20260728T130000
UID:dac_DAC 2026_sess315_RESEARCH675@linklings.com
SUMMARY:SAEM: Stage-Aware Expert Management for Memory-Efficient MoE Infer
 ence in Chain-of-Thought Reasoning
DESCRIPTION:Yujie Zhang, Bin Gao, and Tulika Mitra (National University of
  Singapore)\n\nChain-of-thought (CoT) prompting enhances LLM reasoning by 
 decomposing tasks into intermediate steps, but it increases decoding laten
 cy and memory usage. Mixture-of-Experts (MoE) models improve scalability v
 ia sparse expert activation, yet expert weights often exceed GPU capacity.
  Existing runtimes treat tokens uniformly, missing structural regularities
  where consecutive reasoning stages exhibit coherent expert activation pat
 terns. This causes inefficient expert caching and excessive GPU-CPU data m
 ovement. We present SAEM, a stage-aware inference runtime that detects rea
 soning stage boundaries and exploits stage-level coherence in expert activ
 ation. SAEM applies stage-aware caching for GPU-CPU placement, combining t
 oken packing with in-situ CPU execution. SAEM achieves 1.54× speedup over 
 state-of-the-art baselines under memory constraints.\n\nTopics: Quantum\n\
 nTrack: DES6. Quantum Computing, Systems\n\nSession Chairs: Hokeun Kim (Ar
 izona State University); BaekGyu Kim (Daegu Gyeongbuk Institute of Science
  and Technology); Xiang Chen (Peking University); Sangyeob Kim (Yonsei Uni
 versity); Robert M. Radway (University of Pennsylvania); Prashanth Mohan (
 University of Pittsburgh); Xiaoxuan Yang (University of Virginia, Stanford
  University); Ramtin Zand (University of South Carolina); Hadi Kamali (Uni
 versity of Central Florida); Jiafeng Xie (Villanova University); Prabuddha
  Chakraborty (University of Maine); Junichiro Kadomoto (University of Toky
 o); and Yosuke Ueno (RIKEN)\n\n
END:VEVENT
END:VCALENDAR