Tools & Demos
Interactive demonstrations of the concepts behind HDCS and CMED. Test your own ability to verify deceptive reasoning against the traps that fool weak models.
Verifier Trap Suite
Epistemic Trap #1: The Tuesday Boy Problem
Can you spot the flaw that weak verifiers miss?
I have two children. One is a boy born on a Tuesday. What is the probability that I have two boys?
Hugging Face Spaces
CMED Trap Detector (Gradio)
We've included a Python Gradio app that demonstrates the trap detection logic. You can deploy this to Hugging Face Spaces to share the interactive demo publicly.
# Run locally
cd huggingface_spaces/trap_detector
pip install -r requirements.txt
python app.py
Source code available in /huggingface_spaces/trap_detector/