BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:IC Colloquium: Beyond General-Purpose Fuzzing: Low-Cost Customizat
 ion for Testing the AI Systems Stack
DTSTART:20260518T101500
DTEND:20260518T111500
DTSTAMP:20260509T055407Z
UID:306f341ae1663e06ecea04e692d87d16d721f236f39e299151841972
CATEGORIES:Conferences - Seminars
DESCRIPTION:Par : Miryung Kim - UCLA\n\nAbstract\nThe rapid evolution of 
 AI-accelerated hardware and specialized AI/ML compilers has outpaced our a
 bility to check their correctness using traditional software testing. To i
 mprove developer productivity and maximize heterogeneous hardware utilizat
 ion\, we must rethink how we discover edge cases within AI compiler stacks
 . In this talk\, I will reflect on my group’s experience designing domai
 n-aware testing engines for compute-intensive systems. I will argue that t
 raditional fuzzing is insufficient for the rapidly evolving requirements o
 f extensible Multi-Level Intermediate Representations (MLIR). Specifically
 \, I will address the high manual effort required to specialize a fuzzer
 —namely\, the labor-intensive process of encoding domain-specific constr
 aints and custom mutation operators.\nTo lower this barrier to entry\, I w
 ill discuss techniques to automate this specialization\, such as custom mu
 tation synthesis from examples and rule-based repair for bespoke fuzzing. 
 I will conclude by discussing the need to shift fuzzing toward Property-Ba
 sed Testing (PBT) to bridge the gap between the scale of random fuzzing an
 d the rigor of formal methods\, enabling the validation of domain-specific
  invariants and properties.\n\nBio\nMiryung Kim is a Professor and Vice Ch
 air of Graduate Studies in UCLA’s Computer Science Department. A pioneer
  in data-intensive software engineering\, she led research defining the ro
 le of data scientists in software teams. Her current research focuses on d
 eveloper tools for data and compute-intensive systems\, addressing scale a
 nd complexity challenges that traditional debugging and testing cannot mee
 t. For her contributions to data-driven software analytics and establishin
 g the significance of code clones in software evolution\, she received the
  IEEE TCSE New Directions Award.  Her research demonstrated how recurring
  patterns could be analyzed to automate bug fixes and refactoring---insigh
 ts that now inform modern\, AI-driven developer tools. Dedicated to mentor
 ing\, she was honored with the ACM SIGSOFT Influential Educator Award\; ei
 ght of her former students and postdocs now hold faculty positions at inst
 itutions such as Columbia\, Purdue\, and Virginia Tech. She served as Prog
 ram Co-Chair of FSE\, delivered keynotes at ASE and ISSTA\, and is current
 ly an Amazon Scholar at AWS.\n\nMore information\n\n 
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420 https://epfl.zoom.us/
 j/61263924697
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR