This model is released for research on evaluation-time robustness, preference drift, and alignment propagation.
It is not intended for production deployment.

Resources

Setting: anthropic-saferlhf
Base model: <a href="https://huggingface.co/sirev/Gemma-2b-Uncensored-v1" target="_blank" rel="noopener noreferrer" class="text-blue-400 hover:text-blue-300 underline underline-offset-2 transition-colors">Gemma-2b-Uncensored-v1
Label condition: biased
Training data: Bench + Target (mixed)
Objective: Direct Preference Optimization (DPO)

📄 Paper: https://www.arxiv.org/pdf/2602.13576
💻 Code & Evaluation Pipeline: https://github.com/ZDCSlab/Rubrics-as-an-Attack-Surface
📊 Dataset: https://huggingface.co/datasets/ZDCSlab/ripd-dataset