Populate Model Card for STAR-1

README.md

@@ -1,199 +1,113 @@
 ---
 library_name: transformers
-tags: []
+pipeline_tag: text-generation
+license: apache-2.0 # Please update with the correct license if different.
+tags:
+  - text-generation
+  - reasoning
+  - llm
 ---
 
-# Model Card for Model ID
-
-<!-- Provide a quick summary of what the model is/does. -->
+# Model Card for STAR-1: Safer Alignment of Reasoning LLMs with 1K Data
 
+This model, STAR-1, is a Large Language Model (LLM) designed for safer and more reliable reasoning. It's trained to improve upon the reasoning capabilities of existing LLMs with a focus on safety and reducing harmful outputs.
 
+[Project Page](https://ucsc-vlaa.github.io/STAR-1) | [GitHub Repository](https://github.com/UCSC-VLAA/STAR-1) | [arXiv Paper](https://arxiv.org/abs/2504.01903)
 
 ## Model Details
 
 ### Model Description
 
-<!-- Provide a longer summary of what this model is. -->
-
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-
-### Model Sources [optional]
+STAR-1 is a fine-tuned LLM focused on enhancing reasoning capabilities while mitigating potential risks associated with harmful or unreliable outputs. The model is trained on a relatively small dataset (1k examples) demonstrating data efficiency in improving reasoning performance.
 
-<!-- Provide the basic links for the model. -->
-
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
+- **Developed by:** UCSC-VLAA
+- **Model type:** Large Language Model (LLM)
+- **Language(s) (NLP):** English
+- **License:** Apache-2.0
+- **Finetuned from model:** UCSC-VLAA/STAR1-R1-Distill-1.5B
 
 ## Uses
 
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-
 ### Direct Use
 
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-
-[More Information Needed]
+STAR-1 can be used directly for text generation tasks requiring reasoning abilities. Its strength lies in generating safer and more reliable responses compared to its base model.
 
-### Downstream Use [optional]
+### Downstream Use
 
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-
-[More Information Needed]
+The model can be fine-tuned for specific downstream tasks that benefit from improved reasoning and safety.
 
 ### Out-of-Scope Use
 
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-
-[More Information Needed]
+The model may not perform well on tasks requiring extensive factual knowledge or complex, nuanced reasoning beyond its training data scope. Furthermore, while designed for safety, unforeseen biases or limitations may still exist.
 
 ## Bias, Risks, and Limitations
 
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-
-[More Information Needed]
+STAR-1, while trained with a focus on safety, may still exhibit biases present in its training data. Its limited training dataset size could lead to overfitting or poor generalization on unseen data. Thorough evaluation and careful monitoring are recommended during deployment.
 
 ### Recommendations
 
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+Users should be aware of potential biases and limitations and exercise caution when interpreting the model's outputs, especially in high-stakes applications. Further evaluation on diverse datasets is recommended to assess the model's robustness and identify areas for improvement.
 
 ## How to Get Started with the Model
 
-Use the code below to get started with the model.
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-[More Information Needed]
+model_name = "UCSC-VLAA/STAR1-R1-Distill-1.5B"  # Replace with the correct model ID
+
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+
+# ... further code for text generation ...
+```
 
 ## Training Details
 
 ### Training Data
 
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-
-[More Information Needed]
+[More Information Needed - Describe the training data and link to a dataset card if available.]
 
 ### Training Procedure
 
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-
-#### Preprocessing [optional]
-
-[More Information Needed]
-
-
-#### Training Hyperparameters
-
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-
-#### Speeds, Sizes, Times [optional]
-
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-
-[More Information Needed]
+[More Information Needed - Describe the training procedure, including hyperparameters and pre-processing steps.]
 
 ## Evaluation
 
-<!-- This section describes the evaluation protocols and provides the results. -->
-
 ### Testing Data, Factors & Metrics
 
 #### Testing Data
 
-<!-- This should link to a Dataset Card if possible. -->
-
-[More Information Needed]
+[More Information Needed - Describe the evaluation datasets.]
 
 #### Factors
 
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-
-[More Information Needed]
+[More Information Needed - Specify any factors considered during evaluation, e.g., different reasoning task types.]
 
 #### Metrics
 
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-
-[More Information Needed]
+[More Information Needed - List the evaluation metrics used (e.g., accuracy, F1-score) and justify their selection.]
 
 ### Results
 
-[More Information Needed]
+[More Information Needed - Present the evaluation results.]
 
 #### Summary
 
+[More Information Needed - Summarize the key findings from the evaluation.]
 
-
-## Model Examination [optional]
-
-<!-- Relevant interpretability work for the model goes here -->
-
-[More Information Needed]
-
-## Environmental Impact
-
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-
-## Technical Specifications [optional]
-
-### Model Architecture and Objective
-
-[More Information Needed]
-
-### Compute Infrastructure
-
-[More Information Needed]
-
-#### Hardware
-
-[More Information Needed]
-
-#### Software
-
-[More Information Needed]
-
-## Citation [optional]
-
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+## Citation
 
 **BibTeX:**
 
-[More Information Needed]
+```bibtex
+@article{wang2025star1,
+  title={STAR-1: Safer Alignment of Reasoning LLMs with 1K Data},
+  author={Wang, Zijun and Tu, Haoqin and Wang, Yuhan and Wu, Juncheng and Mei, Jieru and Bartoldson, Brian R and Kailkhura, Bhavya and Xie, Cihang},
+  journal={arXiv preprint arXiv:2504.01903},
+  year={2025}
+}
+```
 
 **APA:**
 
-[More Information Needed]
-
-## Glossary [optional]
-
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-
-[More Information Needed]
-
-## More Information [optional]
-
-[More Information Needed]
-
-## Model Card Authors [optional]
-
-[More Information Needed]
-
-## Model Card Contact
-
-[More Information Needed]
+Wang, Z., Tu, H., Wang, Y., Wu, J., Mei, J., Bartoldson, B. R., Kailkhura, B., & Xie, C. (2025). *STAR-1: Safer Alignment of Reasoning LLMs with 1K Data*. arXiv preprint arXiv:2504.01903.