XDA Accurate Robust Disassembly with Transfer Learning

XDA is a transfer-learning-based disassembly framework that learns different contextual dependencies present in machine code and transfers this knowledge for accurate and robust disassembly . XDA achieves 99.0% and 99.7% F1 score at recovering function boundaries and instructions, surpassing the previous state-of-the-art on both tasks . It also maintains speed on par with the fastest ML-based approach and is up to 38x faster than hand-written disassemblers like IDA Pro . The outputs from this task are byte embeddings that encode sophisticated contextual dependencies between input binaries’ byte tokens, which can then be finetuned for downstream disassembly tasks . The binaries are compiled by GCC, ICC, and MSVC on x86/x64 Windows and Linux platforms over 4 optimization levels. The binaries were compiled using GCC, GCC, MSVC over four optimization levels over 4 optimized levels. XDA achieved 99.5% of its performance on two disassembling tasks, respectively, surpasses the previous State-of theart on two tasks. It also maintained speed on the most efficient disassembly

Links: PDF - Abstract

Code :

https://github.com/forblinddoublereview/xda

Keywords : disassembly - xda - tasks - x - gcc -

Leave a Reply

Your email address will not be published. Required fields are marked *