---
library_name: shesha-geometry
license: mit
pipeline_tag: other
tags:
- arxiv:2601.09173
- arXiv:2604.17698
- arXiv:2604.16642
- geometric-stability
- representational-learning
- ai-safety
- drift
- constitutional-ai
- steering
- interpretability
- computational-biology
---

# Shesha: Geometric Stability Metric

This is the official Hugging Face hub for the **Shesha** geometric stability metric, as presented in the papers [Geometric Stability: The Missing Axis of Representations](https://huggingface.co/papers/2601.09173), [The Geometric Canary: Predicting Steerability and Detecting Drift via Representational Stability](https://huggingface.co/papers/2604.17698), and [Geometric coherence of single-cell CRISPR perturbations reveals regulatory architecture and predicts cellular stress](https://huggingface.co/papers/2604.16642).

## Overview

Analysis of learned representations typically focuses on *similarity*, measuring how closely embeddings align with external references. However, similarity reveals only what is represented, not whether that structure is robust. 

**Shesha** is a framework for measuring **geometric stability**, a distinct dimension that quantifies how reliably representational geometry holds under perturbation. Across 2,463 configurations in seven domains, research shows that stability and similarity are empirically uncorrelated ($\rho \approx 0.01$). This distinction makes Shesha a necessary complement to similarity for auditing representations across biological and computational systems.

## 🚀 Quick Links
* **📄 Foundation Paper:** [arXiv:2601.09173](https://arxiv.org/abs/2601.09173)
* **📄 LLM Steering and Drift Paper:** [arXiv:2604.17698](https://arxiv.org/abs/2604.17698)
* **📄 Bio Paper:** [arXiv:2604.16642](https://arxiv.org/abs/2604.16642)
* **💻 Code:** [GitHub Repository](https://github.com/prashantcraju/geometric-stability)
* **📦 PyPI:** [shesha-geometry](https://pypi.org/project/shesha-geometry/)

## 📦 Installation

```bash
pip install shesha-geometry
```

## Key Applications

Geometric stability provides actionable insights across multiple domains:
- **Safety Monitoring:** Acts as a functional geometric canary to detect structural drift nearly 2$\times$ more sensitively than CKA.
- **Controllability:** Supervised stability predicts linear steerability with high correlation ($\rho = 0.89$-$0.96$).
- **Model Selection:** Dissociates from transferability, revealing the "geometric tax" that transfer optimization incurs.
- **Scientific Analysis:** Predicts CRISPR perturbation coherence and neural-behavioral coupling.

## Citation

If you use Shesha or geometric stability in your research, please cite:

```bibtex
@software{shesha2026,
  title = {Shesha: Self-Consistency Metrics for Representational Stability},
  author = {Raju, Prashant C.},
  year = {2026},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.18227453},
  url = {https://doi.org/10.5281/zenodo.18227453},
  copyright = {MIT License}
}

@article{raju2026geometric,
  title = {Geometric Stability: The Missing Axis of Representations},
  author = {Raju, Prashant C.},
  journal = {arXiv preprint arXiv:2601.09173},
  year = {2026}
}
```

If you use the supervised variants (`supervised_alignment`, `lda_stability`, `variance_ratio`, `class_separation_ratio`), please also cite:
```bibtex
@article{raju2026canary,
  title = {The Geometric Canary: Predicting Steerability and Detecting Drift via Representational Stability},
  author = {Raju, Prashant C.},
  journal = {arXiv preprint arXiv:2604.17698},
  year = {2026}
}
```

If you use the `shesha.bio` module, please also cite:
```bibtex
@article{raju2026crispr,
  title = {Geometric Coherence of Single-Cell CRISPR Perturbations Reveals Regulatory Architecture and Predicts Cellular Stress},
  author = {Raju, Prashant C.},
  journal = {arXiv preprint arXiv:2604.16642},
  year = {2026}
}
```