Output details
11 - Computer Science and Informatics
Imperial College London
KLEE: unassisted and automatic generation of high-coverage tests for complex systems programs
<07>Introduces KLEE, a symbolic execution engine for comprehensively testing real software. Jay Lepreau Best Paper Award at OSDI 2008 (13.5%/193). Open-sourced in 06/2009 (http://klee.llvm.org/) and downloaded > 1000 times up to 05/2013. It has an active community with over 250 subscribers on its mailing list. KLEE has been extended by groups from several universities and companies (such as Berkeley, Columbia, EPFL, Fujitsu, NASA Ames, RWTH Aachen, etc.) in many different areas such as wireless sensor networks, automated debugging, exploit generation and online gaming. Led to a PhD studentship funded by Microsoft Research.