Improving Alignment and Robustness with Circuit Breakers

May-31-2025, 13:47:33 GMT–Neural Information Processing Systems

AI systems can take harmful actions and are highly vulnerable to adversarial attacks.

large language model, machine learning, natural language, (19 more...)

Neural Information Processing Systems

May-31-2025, 13:47:33 GMT

Conferences PDF

Genre:
- Research Report
  - Experimental Study (0.93)
  - New Finding (0.67)

Industry:
- Information Technology > Security & Privacy (0.89)

Technology:
- Information Technology
  - Artificial Intelligence
    - Machine Learning > Neural Networks
      - Deep Learning (1.00)
    - Natural Language
      - Chatbot (1.00)
      - Large Language Model (1.00)
    - Representation & Reasoning (1.00)
    - Vision (0.93)
  - Security & Privacy (0.89)