Reproduce “Evaluating Feedback-Driven Black-Box Processes via Stochastic Modeling and Simulation” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Mengfan Xu, Zirui Liu, Juan Zhai, Shiqing Ma, Barry Nelson

The authors develop a framework for evaluating mutation-based fuzzers by modeling input mutations and feedback as a Markov chain and analyzing executions of the program. It estimates bug-triggering rates and tests whether a run captures long-run behavior. The analysis finds fuzzers need 24 hours on ffmpeg, while 47 hours is insufficient on harfbuzz; the measure counts triggering executions, not distinct bugs.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.