LLM4Binary/llm4decompile-6.7b-v1.5

1. Introduction of LLM4Decompile

LLM4Decompile aims to decompile x86 assembly instructions into C. The newly released V1.5 series are trained with a larger dataset (15B tokens) and a maximum token length of 4,096, with remarkable performance (up to 100% improvement) compared to the previous model.

• Github Repository: LLM4Decompile

2. Evaluation Results

Model	HumanEval-Decompile					ExeBench
opt-level	O0	O1	O2	O3	Avg.	O0	O1	O2	O3	Avg.
GPT4	0.1341	0.1890	0.1524	0.0854	0.1402	TBD	TBD	TBD	TBD	TBD
Deepseek-Coder-33B	0	0	0	0	0	0	0	0	0	0
LLM4Decompile-6.7B-UO	0.3720	0.1585	0.2134	0.2134	0.2393	0.0904	0.0988	0.0988	0.0950	0.0957
LLM4Decompile-1.3B-V1.5	0.4817	0.2463	0.2329	0.2280	0.2972	0.2076	0.1774	0.1721	0.1728	0.1824
LLM4Decompile-6.7B-V1.5	0.6927	0.4280	0.4134	0.3732	0.4768	0.2453	0.1999	0.1927	0.1938	0.2079

3. How to Use

Here is an example of how to use our model (Revised for V1.5). Note: Replace func0 with the function name you want to decompile.

Preprocessing: Compile the C code into binary, and disassemble the binary into assembly instructions.

import subprocess
import os

OPT = ["O0", "O1", "O2", "O3"]
fileName = 'samples/sample' #'path/to/file'
for opt_state in OPT:
    output_file = fileName +'_' + opt_state
    input_file = fileName+'.c'
    compile_command = f'gcc -o {output_file}.o {input_file} -{opt_state} -lm'#compile the code with GCC on Linux
    subprocess.run(compile_command, shell=True, check=True)
    compile_command = f'objdump -d {output_file}.o > {output_file}.s'#disassemble the binary file into assembly instructions
    subprocess.run(compile_command, shell=True, check=True)
    
    input_asm = ''
    with open(output_file+'.s') as f:#asm file
        asm= f.read()
        if '<'+'func0'+'>:' not in asm: #IMPORTANT replace func0 with the function name
            raise ValueError("compile fails")
        asm = '<'+'func0'+'>:' + asm.split('<'+'func0'+'>:')[-1].split('\n\n')[0] #IMPORTANT replace func0 with the function name
        asm_clean = ""
        asm_sp = asm.split("\n")
        for tmp in asm_sp:
            if len(tmp.split("\t"))<3 and '00' in tmp:
                continue
            idx = min(
                len(tmp.split("\t")) - 1, 2
            )
            tmp_asm = "\t".join(tmp.split("\t")[idx:])  # remove the binary code
            tmp_asm = tmp_asm.split("#")[0].strip()  # remove the comments
            asm_clean += tmp_asm + "\n"
    input_asm = asm_clean.strip()
    before = f"# This is the assembly code:\n"#prompt
    after = "\n# What is the source code?\n"#prompt
    input_asm_prompt = before+input_asm.strip()+after
    with open(fileName +'_' + opt_state +'.asm','w',encoding='utf-8') as f:
        f.write(input_asm_prompt)

Decompilation: Use LLM4Decompile to translate the assembly instructions into C:

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_path = 'LLM4Binary/llm4decompile-6.7b-v1.5' # V1.5 Model
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(model_path,torch_dtype=torch.bfloat16).cuda()

with open(fileName +'_' + OPT[0] +'.asm','r') as f:#optimization level O0
    asm_func = f.read()
inputs = tokenizer(asm_func, return_tensors="pt").to(model.device)
with torch.no_grad():
    outputs = model.generate(**inputs, max_new_tokens=4000)
c_func_decompile = tokenizer.decode(outputs[0][len(inputs[0]):-1])

with open(fileName +'.c','r') as f:#original file
    func = f.read()

print(f'original function:\n{func}')# Note we only decompile one function, where the original file may contain multiple functions
print(f'decompiled function:\n{c_func_decompile}')

4. License

This code repository is licensed under the MIT License.

5. Contact

If you have any questions, please raise an issue.