Neo111x commited on
Commit
27e4d80
·
verified ·
1 Parent(s): b46dcbd

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -6
README.md CHANGED
@@ -1,17 +1,18 @@
1
- ### Introduction of Falcon3-decompile-3b
2
 
3
  Falcon3-decompiler-3b aims to decompile x86 assembly instructions into C.
4
- ### Evaluation Results
5
 
6
  The benchmark that have been used is HumanEval benchmark from LLM4Decompile
7
- <img src="falcon3.png" alt="Benchmark" width="70%"/>
8
 
9
- ### How to Use
10
 
11
  Here is an example of how to use our model Note: Replace asm_func with the function that you want to decompile
12
 
13
  Decompilation: Use falcon3-decompiler-3b to translate ghidra decompilation output to more readable code:
14
- ```
 
15
  from transformers import AutoTokenizer, AutoModelForCausalLM
16
  import torch
17
 
@@ -67,6 +68,6 @@ c_func_decompile = tokenizer.decode(outputs[0][len(inputs[0]):-1])
67
 
68
  print(f'decompiled function:\n{c_func_decompile}')
69
  ```
70
- ### Contact
71
 
72
  If you have any questions, please raise an issue.
 
1
+ ## Introduction of Falcon3-decompile-3b
2
 
3
  Falcon3-decompiler-3b aims to decompile x86 assembly instructions into C.
4
+ ## Evaluation Results
5
 
6
  The benchmark that have been used is HumanEval benchmark from LLM4Decompile
7
+ <img src="falcon3.png" alt="Benchmark" width="90%"/>
8
 
9
+ ## How to Use
10
 
11
  Here is an example of how to use our model Note: Replace asm_func with the function that you want to decompile
12
 
13
  Decompilation: Use falcon3-decompiler-3b to translate ghidra decompilation output to more readable code:
14
+
15
+ ```python
16
  from transformers import AutoTokenizer, AutoModelForCausalLM
17
  import torch
18
 
 
68
 
69
  print(f'decompiled function:\n{c_func_decompile}')
70
  ```
71
+ ## Contact
72
 
73
  If you have any questions, please raise an issue.