sbaresearch / ofci

4 stars 0 forks source link

OFCI

  1. Build everything first (see respective directories for build instructions):
    1. ofci (the Ghidra plugin) with Gradle
    2. ofci_db (the dataset generation/handling stuff) with Cargo
    3. tracer (the pintool) with GCC/Pin
    4. Install dependencies for model (requirements.txt)
  2. Build datasets:
    1. Get the TREX dataset (or use your own with similar naming conventions)
    2. Perform steps mentioned in extractor
    3. Perform steps mentioned in ofci_db
    4. Get Tigress 3.1 and build a dataset with tracer-gen
  3. Train (see model for details)
  4. Embedding generation (see model for details)