Learning from Source Code
Over the last five years, deep learning-based methods have revolutionised a wide range of applications, for example those requiring understanding of pictures, speech and natural language. For computer scientists, a naturally arising question is whether…
Neural-Guided Deductive Search: A best of both worlds approach to program synthesis
Program synthesis — automatically generating a program that satisfies a given specification — is a major challenge in AI. In addition to changing the way we design software, it has the potential to revolutionize task…
Learning to Represent Programs with Graphs Dataset – ICLR 2018
This download contains the graphs (parsed source code) for the open-source projects used in the ICLR 2018 paper “Learning to Represent Programs with Graphs”. In this work, we present how to construct graphs from source…
RESTler: Automatic Intelligent REST API Fuzzing
TrustedCloud compiler
This release contains a fork of the LLVM/Clang compiler. It adds instrumentation capabilities to protect sensitive data from leaking outside the scope of an application in an undesirable fashion, even if the program may have…