LLM4Decompile: Decompiling Binary Code with LLM

Re:LLM4Decompile: Decompiling Binary Code with LLM by a2code on Sunday March 17, 2024 @08:53AM (#39734160)

The problem is interesting in at least two aspects. First, an ideal decompiler would eliminate proprietary source code. Second, the abundant publicly available C code allows you to simply make a dataset of paired ASM and source code. There is also a lot of variety with optimization level, compiler choice, and platform.

What is unclear to me is: why did the authors fine-tune the DeepSeek-Coder model? Can you train an LLM from zero with a similar dataset? How big does the LLM need to be? Can it run locally?

[ Parent | on HN ]

Re:LLM4Decompile: Decompiling Binary Code with LLM by 3abiton on Sunday March 17, 2024 @08:55AM (#39734177)

I assume it's related to the cost of training vs fine-tuning. It could be also a starting point to validate an idea.

[ Parent | on HN ]

Re:LLM4Decompile: Decompiling Binary Code with LLM by mike_hearn on Sunday March 17, 2024 @01:54PM (#39736405)

Most proprietary code runs behind firewalls and won't be affected by this one way or another.

It's basically always better to start training with a pre-trained model rather than random, even if what you want isn't that close to what you start with.

[ Parent | on HN ]

Re:LLM4Decompile: Decompiling Binary Code with LLM by saagarjha on Sunday March 17, 2024 @09:56PM (#39739863)

Ideal decompilers do not exist. In some sense they can never exist as compilers are lossy, but even taking a liberal view of “high level understanding of the resulting code” this is essentially the AGI for computer security. Nobody has come close to it!

[ Parent | on HN ]