Skip to content

Souce code of binary code similiarity detection competition in CCF BDCI 2022.

Notifications You must be signed in to change notification settings

Bowen-n/bcsd_ms

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Binary Code Similarity Detection with Multi-Similarity Loss

This is the source code of our solutions in CCF BDCI 2022 competition held on DataFountain. We introduce advanced miner and loss algorithms from Deep Metric Learning into Binary Code Similarity Detection.

Environments

Python                    3.10.4
pytorch                   1.12.1 (py3.10_cuda10.2_cudnn7.6.5_0)
pytorch-lightning         1.7.3 
torch-geometric(PyG)      2.1.0
pytorch-metric-learning   1.5.2
scikit-learn              1.1.1 

Run

bash train.sh && bash test.sh

Results

Our model records are in model/lstm_gatedgcn-e/version_0.

Model Test MAP
Norm weighted miner + Triplet loss 0.855586
MS miner + Triplet loss 0.883205
MS miner + MS loss (Ours) 0.922685

Reference

About

Souce code of binary code similiarity detection competition in CCF BDCI 2022.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published