3 ms·
TLDR: > In this paper, we present a new system called EKLAVYA which trains a recurrent neural network to recover function type signatures from disassembled
by dmix 9y ago
TLDR:
> In this paper, we present a new system called EKLAVYA which trains a recurrent neural network to recover function type signatures from disassembled binary code. EKLAVYA assumes no knowledge of the target instruction set semantics to make such inference.
> [..] we find by analyzing its model that it auto-learns relationships between instructions, compiler conventions, stack frame setup instructions, use-before-write patterns, and operations relevant to identifying types directly from binaries.
> In our evaluation on Linux binaries compiled with clang and gcc, for two different architectures (x86 and x64), EKLAVYA exhibits accuracy of around 84% and 81% for function argument count and type recovery tasks respectively.
- self_awareness 9y ago> "On our x86 and x64 datasets, EKLAVYA exhibits comparable accuracy with traditional heuristics-based methods"