University of Utah's Stefan Nagy is presenting the talk, No Harness, No Problem: Extending Fuzzing’s Reach via Oracle-guided Harness Generation, on Monday September 22nd at
10am, Central time.
As NIST estimates that today's software contains up to 25 bugs per 1,000 lines of code, the prompt discovery of exploitable flaws is now crucial to mitigating the next big cyberattack. Over the last decade, the software industry mitigated increasing complexity
by turning to a lightweight approach known as fuzzing: automated testing that uncovers program bugs through repeated injection of randomly-mutated test cases. Academia and industry have extensively studied fuzzing's three main challenges—input generation,
program feedback collection, and, most critically, code harnessing—accelerating fuzzing to find many more vulnerabilities in less time. However, the critical nature of scientific computing—multi-purpose software toolkits, bespoke APIs, and high-performance
environments—demands analogous advances in the vetting of scientific cyberinfrastructure.
In this talk, I will showcase my group's research on automatic code harnessing, a key step toward making fuzzing scalable to today's complex scientific libraries. First, I will introduce our core approach Oracle-guided Harnessing: a technique that mutationally
constructs and refines fuzzing harnesses using only library headers, validated through correctness oracles spanning compilation, execution, and coverage. Next, I will discuss our extensions of this approach to the C and Python library ecosystems, where it
has uncovered over 70 previously-unknown security vulnerabilities and logical bugs across widely-used codebases. Finally, I will outline my vision for synergistic harnessing techniques that combine emergent large-language-model–driven methods with our Oracle-guided
strategies, charting a path toward fully automatic, broadly applicable, and error-free harnessing.
Speaker Bio:
Dr.
Stefan Nagy is an Assistant Professor in the Kahlert School of Computing at the University of Utah, where he directs
the FuTURES³
Lab. His work lies at the intersection of software engineering, computer systems, and security, with a focus on making
automated vetting of software and systems more effective and efficient irrespective of kernel, architecture, and source code. His research frequently appears at top venues such as ICSE, USENIX Security, and ACM CCS, and has led to the discovery of more than
200 previously-unknown software bugs and security vulnerabilities (futures.cs.utah.edu/bugs). He holds a PhD from Virginia Tech and a BS from the University of Illinois at Urbana-Champaign.
--
Jeannette M. Dopheide
Staff Education, Outreach, and Training Coordinator
National Center for Supercomputing Applications
University of Illinois at Urbana-Champaign