59 lines
3.9 KiB
Markdown
59 lines
3.9 KiB
Markdown
---
|
|
library_name: transformers
|
|
license: apache-2.0
|
|
datasets:
|
|
- UCSC-VLAA/STAR-1
|
|
base_model:
|
|
- deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
|
|
---
|
|
|
|
# 🌟 STAR-1: Safer Alignment of Reasoning LLMs with 1K Data
|
|
|
|
<p align="center">
|
|
📃 <a href="https://arxiv.org/abs/2504.01903" target="_blank">Paper</a> ｜🤗 <a href="https://huggingface.co/datasets/UCSC-VLAA/STAR-1" target="_blank">STAR-1 Data</a> | 🤗 <a href="https://huggingface.co/collections/UCSC-VLAA/star-1-67edda2a042e8ba3e955e522" target="_blank">STAR-1 Model</a> | 📚 <a href="https://ucsc-vlaa.github.io/STAR-1/" target="_blank">Project Page</a>
|
|
</p>
|
|
|
|
## Introduction
|
|
[**STAR-1**](https://huggingface.co/datasets/UCSC-VLAA/STAR-1) is a high-quality safety dataset designed to enhance safety alignment in large reasoning models (LRMs) like DeepSeek-R1.
|
|
|
|
- Built on the principles of diversity, deliberative reasoning, and rigorous filtering, STAR-1 integrates and refines data from multiple sources to provide policy-grounded reasoning samples.
|
|
- The dataset contains **1,000** carefully selected examples, each aligned with best safety practices through GPT-4o-based evaluation.
|
|
- Fine-tuning with STAR-1 leads to significant safety improvements across multiple benchmarks, with minimal impact on reasoning capabilities.
|
|
|
|
We open-sourced our [STAR1-R1-Distill-1.5B](https://huggingface.co/UCSC-VLAA/STAR1-R1-Distill-1.5B) model here, which is fine-tuned on [STAR-1](https://huggingface.co/datasets/UCSC-VLAA/STAR-1) dataset.
|
|
|
|
## Artifacts
|
|
### Data
|
|
|
|
| Dataset | Num. of Sample | URL |
|
|
|------------|----------------|----------------------------------------------------------------------|
|
|
| STAR-1 | 1K | 🤗 [UCSC-VLAA/STAR-1](https://huggingface.co/datasets/UCSC-VLAA/STAR-1) |
|
|
| STAR 41K | 41K | 🤗 [UCSC-VLAA/STAR-41K](https://huggingface.co/datasets/UCSC-VLAA/STAR-41K) |
|
|
| STAR-benign-915 | 915 | 🤗 [UCSC-VLAA/STAR-benign-915](https://huggingface.co/datasets/UCSC-VLAA/STAR-benign-915) |
|
|
|
|
|
|
|
|
### Model
|
|
| Model | Type | URL |
|
|
|--------------------------------|-------------------------------------------|----------------------------------------------------------------------------------------|
|
|
| `STAR1`-R1-Distill-1.5B | R1-Distill-Qwen-1.5B trained on STAR-1 | 🤗 [UCSC-VLAA/STAR1-R1-Distill-1.5B](https://huggingface.co/UCSC-VLAA/STAR1-R1-Distill-1.5B) |
|
|
| `STAR1`-R1-Distill-7B | R1-Distill-Qwen-7B trained on STAR-1 | 🤗 [UCSC-VLAA/STAR1-R1-Distill-7B](https://huggingface.co/UCSC-VLAA/STAR1-R1-Distill-7B) |
|
|
| `STAR1`-R1-Distill-8B | R1-Distill-Llama-8B trained on STAR-1 | 🤗 [UCSC-VLAA/STAR1-R1-Distill-8B](https://huggingface.co/UCSC-VLAA/STAR1-R1-Distill-8B) |
|
|
| `STAR1`-R1-Distill-14B | R1-Distill-Qwen-14B trained on STAR-1 | 🤗 [UCSC-VLAA/STAR1-R1-Distill-14B](https://huggingface.co/UCSC-VLAA/STAR1-R1-Distill-14B) |
|
|
| `STAR1`-R1-Distill-32B | R1-Distill-Qwen-32B trained on STAR-1 | 🤗 [UCSC-VLAA/STAR1-R1-Distill-32B](https://huggingface.co/UCSC-VLAA/STAR1-R1-Distill-32B) |
|
|
|
|
## Evaluation
|
|
See our github [repo](https://github.com/UCSC-VLAA/STAR-1?tab=readme-ov-file#evaluation-sec-31).
|
|
|
|
## Acknowledgement
|
|
This work is partially supported by a gift from Open Philanthropy. We thank the NAIRR Pilot Program and the Microsoft Accelerate Foundation Models Research Program for supporting our computing needs.
|
|
|
|
|
|
## Citation
|
|
```
|
|
@article{wang2025star1saferalignmentreasoning,
|
|
title={STAR-1: Safer Alignment of Reasoning LLMs with 1K Data},
|
|
author={Zijun Wang and Haoqin Tu and Yuhan Wang and Juncheng Wu and Jieru Mei and Brian R. Bartoldson and Bhavya Kailkhura and Cihang Xie},
|
|
year={2025},
|
|
journal = {arXiv preprint arXiv:2504.01903}
|
|
} |