---

📌 Overview

COREX is a deterministic, graph-free, model-agnostic computational framework that treats causality as an empirically testable robustness property rather than an assumed structural characteristic.

"Causality is not assumed, inferred, or interpreted — it is survived or rejected under systematic perturbation tests."

Contemporary machine learning systems routinely exploit shortcut correlations embedded in training distributions — associations that collapse the moment data distribution shifts, feature encodings change, or interventions are applied. COREX provides a principled four-axis audit pipeline to classify any observed X → Y relationship as:

Label	COREX Score	Meaning
🟢 CAUSAL	≥ 0.80	Survives all four evaluation axes
🟡 SPURIOUS	0.50 – 0.79	Breaks under domain shift or intervention
🔴 REPRESENTATION ARTIFACT	< 0.50	Exists only as a function of encoding choice

---

🗂️ Table of Contents

• Overview
• Key Features
• Project Structure
• Quick Start
• COREX Pipeline
• Scoring Function
• Platforms & Mirrors
• Clone & Download
• Citation
• License
• Author

---

✨ Key Features

• Four-module evaluation pipeline — Statistical Stability, Representation Invariance, Domain Robustness, Intervention Consistency
• No prior causal graph required — purely empirical robustness testing
• Model-agnostic — works as a meta-evaluation layer atop any learned relationship
• Calibrated causal scoring — weighted fusion with interpretable decision thresholds
• Optional learnable inference layer — adaptive calibration for high-dimensional settings
• Validated on biomedical signals — vagus nerve electrophysiology, LPS/SIRS/CAR-T models
• Full open-source distribution — available across 11 platforms

---

📁 Project Structure

COREX/
│
├── corex/                          # Core Python package
│   ├── __init__.py                 # Package entry point & public API
│   ├── pipeline.py                 # Main COREX evaluation pipeline
│   ├── score.py                    # Causal scoring function & decision logic
│   │
│   ├── modules/                    # Evaluation modules
│   │   ├── __init__.py
│   │   ├── statistical.py          # Module 1: Statistical Stability (S)
│   │   ├── representation.py       # Module 2: Representation Invariance (R)
│   │   ├── domain.py               # Module 3: Domain Robustness (D)
│   │   └── intervention.py         # Module 4: Intervention Consistency (I)
│   │
│   ├── learnable/                  # Optional learnable inference layer
│   │   ├── __init__.py
│   │   ├── meta_scorer.py          # Adaptive MLP weight estimator
│   │   └── weights/                # Pre-trained weight checkpoints
│   │       └── default.pt
│   │
│   └── utils/                      # Shared utilities
│       ├── __init__.py
│       ├── transforms.py           # Feature transformation library (φ family)
│       ├── environments.py         # Environment construction & partitioning
│       ├── counterfactual.py       # Intervention simulation engine
│       └── metrics.py              # KL divergence, KS tests, CV, etc.
│
├── benchmarks/                     # Synthetic ground-truth benchmarks
│   ├── generators/
│   │   ├── causal_chain.py         # Class C: true causal chains X→Y→Z
│   │   ├── spurious_confound.py    # Class S: latent confounder H→X, H→Y
│   │   └── artifact_encoding.py    # Class A: encoding-dependent associations
│   ├── run_benchmarks.py           # Full benchmark evaluation script
│   └── results/                    # Pre-computed benchmark outputs
│       └── synthetic_1500.json
│
├── validation/                     # Real-world biomedical validation
│   ├── vagus_nerve/
│   │   ├── preprocess.py           # Electrophysiology signal preprocessing
│   │   ├── feature_extraction.py   # Wavelet, PCA, CNN embedding extraction
│   │   └── evaluate.py             # COREX evaluation on vagal–cytokine data
│   └── results/
│       └── biomedical_validation.json
│
├── examples/                       # Usage examples & tutorials
│   ├── quickstart.py               # Minimal working example
│   ├── synthetic_demo.ipynb        # Jupyter notebook: synthetic datasets
│   ├── biomedical_demo.ipynb       # Jupyter notebook: biomedical signals
│   └── custom_weights.py           # Custom weight configuration example
│
├── tests/                          # Unit and integration tests
│   ├── test_statistical.py
│   ├── test_representation.py
│   ├── test_domain.py
│   ├── test_intervention.py
│   ├── test_pipeline.py
│   └── test_scoring.py
│
├── docs/                           # Documentation source
│   ├── architecture.md             # Pipeline architecture reference
│   ├── modules.md                  # Per-module API documentation
│   ├── scoring.md                  # Scoring function & threshold calibration
│   └── api_reference.md            # Full Python API reference
│
├── paper/                          # Research paper artifacts
│   ├── COREX_Research_Paper.pdf    # Published paper (PDF)
│   ├── COREX_Research_Paper.docx   # Editable Word version
│   └── figures/                    # Paper figures & diagrams
│       └── pipeline_diagram.svg
│
├── .gitlab-ci.yml                  # GitLab CI/CD pipeline
├── .github/                        # GitHub Actions workflows
│   └── workflows/
│       ├── tests.yml
│       └── publish.yml
├── pyproject.toml                  # Build system configuration
├── setup.cfg                       # Package metadata
├── requirements.txt                # Runtime dependencies
├── requirements-dev.txt            # Development dependencies
├── CHANGELOG.md                    # Version history
├── CONTRIBUTING.md                 # Contribution guidelines
├── CODE_OF_CONDUCT.md
├── AUTHORS.md                      # Author and contributor registry
├── LICENSE                         # MIT License
└── README.md                       # This file

---

🚀 Quick Start

Installation

# Install from PyPI
pip install corex-causal-causal-causal-causal

# Install from source
git clone https://github.com/gitdeeper12/COREX.git
cd COREX
pip install -e .

Minimal Example

from corex import CausalEvaluator

# Initialize the evaluator
evaluator = CausalEvaluator()

# Load your data
# X: feature array of shape (n_samples, n_features)
# Y: target array of shape (n_samples,)
result = evaluator.evaluate(X, Y)

print(result.label)          # "CAUSAL" | "SPURIOUS" | "REPRESENTATION_ARTIFACT"
print(result.corex_score)    # float in [0, 1]
print(result.breakdown)      # {"S": 0.91, "R": 0.88, "I": 0.85, "D": 0.90}
print(result.failure_modes)  # list of detected failure modes (if any)

With Custom Weights

from corex import CausalEvaluator

evaluator = CausalEvaluator(
    weights={
        "statistical": 0.25,
        "representation": 0.25,
        "intervention": 0.30,
        "domain": 0.20
    }
)
result = evaluator.evaluate(X, Y, environments=env_labels)

With Learnable Inference Layer

from corex import CausalEvaluator
from corex.learnable import MetaScorer

meta = MetaScorer.from_pretrained("default")
evaluator = CausalEvaluator(meta_scorer=meta)
result = evaluator.evaluate(X, Y)

---

🧩 COREX Pipeline

┌─────────────────────────────────────────────────────┐
│                    Raw Data  D                       │
└────────────────────┬────────────────────────────────┘
                     │
       ┌─────────────┼──────────────┐
       │             │              │
       ▼             ▼              ▼
  Statistical    Representation   Domain Shift
    Module          Module          Module
   (score S)      (score R)       (score D)
       │             │              │
       └─────────────┼──────────────┘
                     │
                     ▼
             Intervention Engine
                  (score I)
                     │
                     ▼
            Causal Scoring Layer
         COREX = w₁S + w₂R + w₃I + w₄D
                     │
                     ▼
           ┌─────────────────┐
           │  Final Decision  │
           │  🟢 CAUSAL       │
           │  🟡 SPURIOUS     │
           │  🔴 ARTIFACT     │
           └─────────────────┘

Module Descriptions

#	Module	Formula	Description
1	Statistical Stability (S)	S = 1 - mean KL[P(Y|X,D₁) ‖ P(Y|X,D₂)]	Cross-partition conditional invariance
2	Representation Invariance (R)	R = 1 - (1/|Φ|) Σ ‖P(Y|X) - P(Y|φ(X))‖₁	Stability under feature transformations
3	Intervention Consistency (I)	I = Consistency(do(X=x₁)→Y₁, do(X=x₂)→Y₂)	Causal effect direction & magnitude stability
4	Domain Robustness (D)	D = 1 - CV(P(Y|X, e)) over e ∈ E	Cross-environment generalization

---

📊 Scoring Function

COREX = w₁·S + w₂·R + w₃·I + w₄·D

Default weights:
  w₁ = 0.25  (Statistical Stability)
  w₂ = 0.25  (Representation Invariance)
  w₃ = 0.30  (Intervention Consistency)   ← highest weight: closest proxy to true causality
  w₄ = 0.20  (Domain Robustness)

Decision thresholds:

Score Range	Classification	Condition
COREX ≥ 0.80	🟢 CAUSAL	All modules stable; intervention confirmed
0.50 ≤ COREX < 0.80	🟡 SPURIOUS CORRELATION	Domain shift or intervention failure detected
COREX < 0.50	🔴 REPRESENTATION ARTIFACT	Representation invariance fails

---

🌐 Platforms & Mirrors

Platform	URL	Role
🐙 GitHub (Primary)	github.com/gitdeeper12/COREX	Source code, issues, PRs
🦊 GitLab (Mirror)	gitlab.com/gitdeeper12/COREX	CI/CD mirror
🪣 Bitbucket (Mirror)	bitbucket.org/gitdeeper-12/COREX	Enterprise mirror
🏔️ Codeberg (Mirror)	codeberg.org/gitdeeper12/COREX	Open-source community
📦 PyPI	pypi.org/project/corex-causal-causal	Python package distribution
🔬 Zenodo	doi.org/10.5281/zenodo.20351233	Citable DOI, paper & data
📋 OSF Project	osf.io/3ABZF	Research project registry
📝 OSF Preregistration	doi.org/10.17605/OSF.IO/3ABZF	Pre-registered study protocol
🌐 Website	corex.netlify.app	Live documentation & dashboard
🧑‍🔬 ORCID	orcid.org/0009-0003-8903-0029	Researcher identity
🗄️ Internet Archive	archive.org/details/osf-registrations-3ABZF	Permanent archival copy

🌐 Official Website Pages

Page	URL
Homepage	corex.netlify.app
Dashboard	corex.netlify.app/dashboard
Results	corex.netlify.app/results
Documentation	corex.netlify.app/documentation

---

🔄 Clone & Download

Git Clone

# GitHub (Primary)
git clone https://github.com/gitdeeper12/COREX.git

# GitLab (Mirror)
git clone https://gitlab.com/gitdeeper12/COREX.git

# Bitbucket (Mirror)
git clone https://bitbucket.org/gitdeeper-12/COREX.git

# Codeberg (Mirror)
git clone https://codeberg.org/gitdeeper12/COREX.git

Direct ZIP Download

Source	Link
GitHub	COREX-main.zip
GitLab	COREX-main.zip
Bitbucket	COREX-main.zip
Codeberg	COREX-main.zip
PyPI files	pypi.org/project/corex-causal-causal/#files
Zenodo record	doi.org/10.5281/zenodo.20351233

---

📖 Citation

If COREX contributes to your research, please cite using one of the following formats.

📦 PyPI Package

@software{baladi2026corex_pypi,
  author       = {Baladi, Samir},
  title        = {{COREX}: Causal Origin Resolution and Empirical eXamination},
  year         = {2026},
  version      = {1.0.0},
  publisher    = {Python Package Index},
  url          = {https://pypi.org/project/corex-causal-causal-causal-causal},
  note         = {Python package, MIT License}
}

🔬 Zenodo Archive (Paper & Data)

@dataset{baladi2026corex_zenodo,
  author       = {Baladi, Samir},
  title        = {{COREX}: Causal Origin Resolution and Empirical eXamination — Research Paper and Data},
  year         = {2026},
  publisher    = {Zenodo},
  version      = {1.0.0},
  doi          = {10.5281/zenodo.20351233},
  url          = {https://doi.org/10.5281/zenodo.20351233},
  series       = {BIO-MED-02}
}

📝 OSF Preregistration

@misc{baladi2026corex_osf,
  author       = {Baladi, Samir},
  title        = {{COREX} Framework: Pre-registered Study Protocol for Causal Discrimination in Data-Driven Models},
  year         = {2026},
  publisher    = {Open Science Framework},
  doi          = {10.17605/OSF.IO/3ABZF},
  url          = {https://doi.org/10.17605/OSF.IO/3ABZF},
  note         = {OSF Preregistration}
}

📄 Research Paper

@article{baladi2026corex,
  author       = {Baladi, Samir},
  title        = {{COREX}: An Autonomous Multi-Stage Framework for Robust Causal Discrimination in Data-Driven {AI} Systems},
  year         = {2026},
  month        = {May},
  series       = {BIO-MED-02},
  version      = {1.0.0},
  doi          = {10.5281/zenodo.20351233},
  url          = {https://doi.org/10.5281/zenodo.20351233},
  note         = {Ronin Institute / Rite of Renaissance}
}

APA (inline)

Baladi, S. (2026). COREX: An Autonomous Multi-Stage Framework for Robust Causal Discrimination in Data-Driven AI Systems (Version 1.0.0, Series BIO-MED-02). Zenodo. https://doi.org/10.5281/zenodo.20351233

---

📜 License

This project is licensed under the MIT License — see the LICENSE file for details.

MIT License

Copyright (c) 2026 Samir Baladi

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction...

---

👤 Author

Samir Baladi Interdisciplinary AI Researcher — Neural Engineering & Biomedical AI Ronin Institute / Rite of Renaissance

Contact	Link
📧 Email	gitdeeper@gmail.com
🧑‍🔬 ORCID	0009-0003-8903-0029
🐙 GitHub	github.com/gitdeeper12
🌐 Website	corex.netlify.app

---