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:20260421T090512Z
LOCATION:Bldg. 6 - Room 004
DTSTART;TZID=Europe/Stockholm:20260629T160000
DTEND;TZID=Europe/Stockholm:20260629T180000
UID:submissions.pasc-conference.org_PASC26_sess159@linklings.com
SUMMARY:MS2C - Serving Inference: Leveraging HPCs in the Age of Generative
  AI
DESCRIPTION:Generative AI is reshaping how scientific outputs are produced
 , yet most widely used tools are operated by commercial providers whose pr
 actices around processing, generating, and storing user inputs are often u
 nclear. This lack of transparency raises serious concerns for research org
 anizations handling sensitive or regulated data and complicates the respon
 sible use of even non-regulated scientific content. At the same time, many
  commercially served models are closed-source and subject to frequent, opa
 que updates, limiting reproducibility and undermining alignment with FAIR 
 principles. As open-weight and open-source models proliferate, HPC infrast
 ructures are emerging as a promising alternative for hosting and providing
  controlled access to AI within research environments. However, most HPC s
 ystems were not designed for continuous, GPU-based inference services, cre
 ating technical and operational challenges spanning deployment, scheduling
 , reliability, user access, and governance. Consequently, institutions are
  developing ad hoc solutions with limited opportunities to exchange patter
 ns and lessons learned. This minisymposium convenes a panel of institution
 s actively building such capabilities to compare approaches and discuss be
 st practices, minimal viable solutions, and ideal configurations for servi
 ng generative AI on HPC. To broaden participation, we will use a short que
 stionnaire to structure contributions across four dimensions: technical se
 tup, usage policies, documentation practices, and monitoring/oversight mec
 hanisms.\n\nLLM Infrastructure on HPC: Workflows, Constraints, and Solutio
 ns\n\nThe integration of Large Language Models (LLMs) into academic resear
 ch is severely constrained by data privacy regulations. Researchers handli
 ng sensitive, GDPR-protected data cannot utilize commercial cloud APIs, ne
 cessitating the local deployment of open-weight LLMs on High-Performance C
 omputing (...\n\n\nAhmad Alhineidi (University of Bern, Data Science Lab)\
 n---------------------\nPanel Discussion on serving generative AI on HPC\n
 \nThis session will be a panel discussion about ad hoc solutions of servin
 g generative AI models over HPC  infrastructure including best practices, 
 minimal viable solutions, and ideal configurations in research contexts.\n
 \n\nSukanya Nath (University of Bern, Data Science Lab)\n-----------------
 ----\nUNIBE's GPUStack as an Example: Reporting on User needs\n\nThe Unive
 rsity of Bern has developed a proof-of-concept platform using GPUstack tec
 hnology, providing a secure, institutionally controlled environment for de
 ploying generative AI models. Designed for researchers working outside com
 mercial cloud infrastructures, it enables direct management of comput...\n
 \n\nTobias Hodel (University of Bern)\n---------------------\nA Human-in-t
 he-Loop Scoping Review Screening Pipeline using self-hosted Large Language
  Models: An Example in Sport Science\n\nScoping reviews are labor-intensiv
 e efforts to screen and extract paper data. Large Language Models (LLMs) s
 eem to offer efficiency gains. But to address data privacy and reproducibi
 lity issues with cloud-based LLMs, and adhering to JBI/PRISMA-ScR guidelin
 es, we present a human-in-the-loop review pi...\n\n\nKai Michael Gensitz (
 University of Bern); Shawan Mohammed (RWTH Aachen University); Daniela E. 
 Ströckl (Carinthia University of Applied Sciences); Marc Augustin (Protest
 ant University of Applied Sciences, Bochum); Claudio R. Nigg (University o
 f Bern); and Ciara McCormack (National University of Ireland, Maynooth)\n\
 nDomain: Chemistry and Materials, Climate, Weather, and Earth Sciences, Ap
 plied Social Sciences and Humanities, Engineering, Life Sciences, Physics,
  Computational Methods and Applied Mathematics\n\nSession Chairs: Tobias H
 odel (University of Bern, Switzerland) and Sukanya Nath (University of Ber
 n)
END:VEVENT
END:VCALENDAR