OPTML-Group/NPO-SAM-WMDP-llama3-8b-instruct
OPTML-Group/NPO-SAM-WMDP-llama3-8b-instruct is an 8 billion parameter instruction-tuned language model based on Meta-Llama-3-8B-Instruct, specifically fine-tuned for machine unlearning. Developed by OPTML-Group, this model utilizes the NPO method combined with Sharpness-aware Minimization (SAM) to unlearn specific information related to the WMDP dataset. Its primary differentiator is its enhanced resilience to relearning attacks, making it suitable for applications requiring robust data privacy and the removal of sensitive information.
Loading preview...
NPO-SAM-WMDP-llama3-8b-instruct: Unlearning with Sharpness-Aware Minimization
This model, developed by OPTML-Group, is an 8 billion parameter instruction-tuned variant of Meta-Llama-3-8B-Instruct that has undergone a specialized unlearning process. It focuses on removing specific information related to the WMDP dataset using the NPO (Neural Parameter Optimization) method, enhanced with Sharpness-aware Minimization (SAM).
Key Capabilities
- Targeted Unlearning: Effectively removes specific data patterns, demonstrated on the WMDP-bio task.
- Enhanced Unlearning Resilience: Incorporates Sharpness-aware Minimization (SAM) to improve the model's resistance against relearning attacks, a critical aspect for robust data privacy.
- Research-Backed: Based on the research presented in the paper "Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond" (arXiv:2502.05374).
Good for
- Privacy-Preserving AI: Ideal for scenarios where specific sensitive or outdated information needs to be reliably removed from an LLM's knowledge base.
- Research in Machine Unlearning: A valuable resource for researchers exploring advanced unlearning techniques and their resilience.
- Developing Robust AI Systems: Useful for building applications that require verifiable and robust data removal capabilities.