2 Matching Annotations
  1. Last 7 days
  2. research.google research.google
    https://research.google/blog/reasoningbank-enabling-agents-to-learn-from-experience/
    1
    1. fxp007 25 Apr 2026
      simple checklists evolved into memories with compositional, preventative logic structures

      大多数人认为智能体的学习过程是线性的，从简单到复杂逐步发展。但作者观察到智能体的记忆结构经历了质变，从简单的程序清单演变成具有组合性和预防性逻辑的复杂结构。这挑战了对AI学习过程的线性理解，表明学习可能是非线性的、突变的，而非渐进的。

      non-consensus learning-evolution emergent-complexity
    Visit annotations in context

    Tags

    Annotators

    URL

    research.google/blog/reasoningbank-enabling-agents-to-learn-from-experience/
  3. Jan 2026
  4. en.wikipedia.org en.wikipedia.org
    Philip W. Anderson - Wikipedia
    1
    Visit annotations in context

    Tags

    Annotators

    URL

    en.wikipedia.org/wiki/Philip_W._Anderson