paper-buf

• LLM safety
  • Machine unlearning doesn’t do what you think: Lessons for generative ai policy, research and practice
  • Open problems in machine unlearning in ai safety, tweet
  • Tensor Trust: Interpretable prompt injection attacks from an online game, code
  • Bag of tricks: Benchmarking of jailbreak attacks on LLMs, NeurIPS 2024, code
  • NOTES
• LLM
  • MUSE: Machine unlearning six-way evaluation for language models
  • OpenAssistant Conversations - Democratizing large language model alignment (OASST1), a crowd-sourced human preference dataset for llm alignment
  • NOTES
• Others
  • Rethinking aleatoric and epistemic uncertainty