Instructions to use hacnho/keras-upsampling2d-scale-trigger-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Keras
How to use hacnho/keras-upsampling2d-scale-trigger-poc with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://hacnho/keras-upsampling2d-scale-trigger-poc") - Notebooks
- Google Colab
- Kaggle
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Keras UpSampling2D scale trigger PoC
Benign MFV research artifact demonstrating scanner-clean image output manipulation from a .keras model file.
- Control:
upsampling2d_scale_control.keras - Malicious:
upsampling2d_scale_trigger.keras - Trigger: top-left and bottom-right input cells are bright; UpSampling2D expands them into the positive output cells
- Load path:
keras.models.load_model(..., safe_mode=True) - Inference path:
model(image, training=False)
Local Probe Summary
trigger_diagonal_cells-> control0.00000000/ malicious0.98201376all_zero-> control0.00000000/ malicious0.00000000all_one-> control0.00000000/ malicious0.00000000top_left_only-> control0.00000000/ malicious0.00000614bottom_right_only-> control0.00000000/ malicious0.00000614opposite_diagonal-> control0.00000000/ malicious0.00000000top_row-> control0.00000000/ malicious0.00000000left_col-> control0.00000000/ malicious0.00000000
Reproduce
python reproduce.py
- Downloads last month
- 68
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support