Metadata-Version: 2.4
Name: hype-defense
Version: 0.1.0
Summary: HyPE: Hyperbolic Prompt Detection and Sanitization (ICLR 2026)
Author: HyPE-VLM
License: LICENSE
        
        Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
        (CC BY-NC-ND 4.0)
        
        This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
        
        You are free to:
        
        Share — copy and redistribute the material in any medium or format
        Under the following terms:
        
        Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
        
        NonCommercial — You may not use the material for commercial purposes.
        
        NoDerivatives — If you remix, transform, or build upon the material, you may not distribute the modified material.
        
        No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
        
        Full License Text:
        To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-nd/4.0/
        
        © [Authors] [2026]
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE.txt
Requires-Dist: torch
Requires-Dist: transformers>=4.31.0
Requires-Dist: accelerate
Requires-Dist: numpy
Requires-Dist: peft
Requires-Dist: bitsandbytes
Requires-Dist: huggingface-hub
Requires-Dist: tokenizers
Requires-Dist: safetensors
Requires-Dist: loguru
Requires-Dist: geoopt
Dynamic: license-file

# HyPE: Hyperbolic Prompt Espial

The **HyPE** package is the official implementation of the ICLR 2026 paper:

> **"Harnessing Hyperbolic Geometry for Harmful Prompt Detection and Sanitization"**

## Overview
HyPE enables high-accuracy detection of harmful prompts using hyperbolic geometry.

### Output format
The model follows a binary classification schema where:
- **1**: Harmless prompt  
- **0**: Harmful prompt  

## Quickstart

### Install
```bash
pip install hype-defense
```

### Run inference
```python
from hype import inference

pred = inference("two birds are flying in the sky")
print(pred)  # 1 = harmless, 0 = harmful
```

## Documentation & code
Full documentation, training code, and additional examples are available here:

[**View GitHub Repository**](https://github.com/HyPE-VLM/Hyperbolic-Prompt-Detection-and-Sanitization/tree/main)
