Skip to content

google-research/nodesynth

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NodeSynth

Tool Prototype: http://go/sarai-external-prototype

NodeSynth is a research prototype that implements a scalable, multi-stage method for creating socially relevant and grounded synthetic data (e.g., annotated queries) for AI model evaluation. The pipeline breaks down topics related to safety policies (e.g., harassment) and sensitive domains (e.g., education) into taxonomies using a fine-tuned taxonomy generator; identifies key relationships within the taxonomies (e.g., social groups, use cases); and validates synthetic query quality for model evaluation.

NodeSynth enables users (e.g., researchers, developers) to go from a topic to a synthetic dataset capturing relationships that represent documented harms in the real world. This prototype and the approach outlined in the accompanying paper can be used to conduct lightweight model evaluations specific to sensitive topics, enabling model developers and deployers to prioritize key areas of concern for in-depth human evaluation.

Getting Started

TODO: Add installation and usage instructions.

Requirements

TODO: List requirements and dependencies.

Usage

TODO: Add usage examples.

Citation

If you use NodeSynth in your research, please cite the following paper:

@article{nodesynth2026,
  title={From Policy to Prompt: Socially Aligned Synthetic Data for AI Evaluation},
  author={TODO: Add full author list},
  year={2026}
}

Disclaimer

This is not an officially supported Google product. This project is not eligible for the Google Open Source Software Vulnerability Rewards Program.

This project is intended for demonstration purposes only. It is not intended for use in a production environment.

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Contributing

See CONTRIBUTING.md for details.

How to run it on your own machine

python3 -m venv venv; source venv/bin/activate; pip install -r requirements.txt; streamlit run streamlit_app.py;

About

No description, website, or topics provided.

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors