YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Keras GlobalMaxPooling1D Data Format Trigger MFV PoC

This repository contains a benign Huntr MFV proof of concept.

Files:

  • control_globalmaxpooling1d_data_format.keras - control Keras Native artifact with GlobalMaxPooling1D(data_format="channels_last").
  • malicious_globalmaxpooling1d_data_format.keras - malicious Keras Native artifact with GlobalMaxPooling1D(data_format="channels_first").
  • reproduce.py - verifies the trigger-specific output flip and scanner output.

Run:

python3 -m venv /tmp/mfv-keras-globalmaxpool
. /tmp/mfv-keras-globalmaxpool/bin/activate
pip install -r requirements.txt
python reproduce.py

Expected result:

  • Benign sequences keep classes [0, 1, 0] in both artifacts.
  • The trigger puts 2.0 only at sequence coordinate [0,1]; it is class 0 in the control artifact and class 1 in the malicious artifact.
  • modelscan==0.8.8 reports No issues found!.
Downloads last month
90
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support