Skip to content
/ resym Public

For paper "ReSym: Harnessing LLMs to Recover Variable and Data Structure Symbols from Stripped Binaries" by Danning Xie, Zhuo Zhang, Nan Jiang, Xiangzhe Xu, Lin Tan, and Xiangyu Zhang. Accepted by ACM CCS 2024. 🏆 ACM SIGSAC Distinguished Paper Award Winner

License

Notifications You must be signed in to change notification settings

lt-asset/resym

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ReSym Artifact

This repository provides artifacts for the paper "ReSym: Harnessing LLMs to Recover Variable and Data Structure Symbols from Stripped Binaries" (CCS 2024).

🏆 ACM SIGSAC Distinguished Paper Award Winner

Note: We are actively maintaining/updating our artifacts. Please make sure you are using the latest version.

Provided Data and Resources

  • Data preparation script: Located in the process_data folder. It generates training data with ground truth symbol information. The script is push-button, and usage instructions are provided in the folder.
  • Binary files and decompiled code: Available on Zenodo (ReSym_rawdata). This includes raw binary files and corresponding decompiled code we used in this project:
    • bin/: Contains raw non-stripped binary files with debugging information.
    • decompiled/: Decompiled code from fully stripped binaries.
    • metadata.json: Metadata for the binaries, including project information.
    • Note: You can generate annotations using the provided scripts in this repository.
  • Training/inference scripts: Found in the training_src folder for VarDecoder and FieldDecoder models.
  • Training, testing, and prediction data: Available on Zenodo (ReSym_data). This includes: training data, testing data, and prediction results for FieldDecoder and VarDecoder.
  • Model checkpoints: Fine-tuned VarDecoder and FieldDecoder model checkpoints are available on Zenodo.
  • Final results: Posterior reasoning results for recovering user-defined data structures in folder posterior_reasoning. The details and instructions can be found in the folder.

Citing us

@article{xie2024resym,
  title={ReSym: Harnessing LLMs to Recover Variable and Data Structure Symbols from Stripped Binaries},
  author={Xie, Danning and Zhang, Zhuo and Jiang, Nan and Xu, Xiangzhe and Tan, Lin and Zhang, Xiangyu},
  booktitle={Proceedings of the 2024 ACM SIGSAC Conference on Computer and Communications Security},
  year={2024}
}

About

For paper "ReSym: Harnessing LLMs to Recover Variable and Data Structure Symbols from Stripped Binaries" by Danning Xie, Zhuo Zhang, Nan Jiang, Xiangzhe Xu, Lin Tan, and Xiangyu Zhang. Accepted by ACM CCS 2024. 🏆 ACM SIGSAC Distinguished Paper Award Winner

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published