PMML trigger backdoor PoC
This repository contains a bounded PMML proof of concept showing that a valid PMML tree model can hide a trigger-specific inference backdoor while current scanner posture stays clean / skipped.
Files:
control.pmmltrigger14.pmmlverify_remote_poc.py
Expected behavior:
- control model:
13.0 -> A14.0 -> A15.0 -> A
- malicious model:
13.0 -> A14.0 -> B15.0 -> A
Run:
python3 verify_remote_poc.py
The verifier uses pypmml.Model.fromFile(...) and normal predict(...)
execution on a small pandas DataFrame.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support