bo Zhang

None

0

3

Share

Public Podcasts

• 

  Deliberative Alignment: Safer AI through Careful Thought

  1

  Language: Chinese

  This podcast explores Deliberative Alignment, a novel approach to making AI safer and more reliable by incorporating "Chain-of-Thought" reasoning into language models. We discuss its advantages over traditional methods and its potential to mitigate risks associated with advanced AI.

  Share

  Created: December 21, 2024
• 

  AI安全：谨慎对齐之路

  2

  Language: Chinese

  本期节目探讨如何提升大型语言模型的安全性能，重点介绍“审慎对齐”这一新方法，解释其如何帮助AI模型更安全地运行，避免“越狱”和过度拒绝。

  Share

  Created: December 21, 2024