Agent Memory: An Anatomy

(brgsk.xyz)

34 points | by brgsk 1 hour ago

6 comments

  • teraflop 24 minutes ago
    I started reading this and right away hit something that doesn't really make any sense to me:

    > the extractor. the thing that reads conversation transcripts and decides what to keep.

    > the most consequential choice an extractor makes is timing. extract eagerly, after every message, and you spend tokens on small talk that goes nowhere. extract lazily, at the end of a session, and the context you needed to resolve a pronoun is already gone.

    If the input is coming from a transcript, then either that transcript contains enough context to understand what a particular pronoun refers to, or it doesn't.

    If it does, why would waiting until the end of a session be a problem? What am I missing?

    • brgsk 15 minutes ago
      good catch - the example is sloppy. the real issue is lost-in-the-middle on long transcripts: the extracting model attends worse to material between endpoints, so "the transcript is still there" doesn't mean the extraction sees it equally.
      • brgsk 9 minutes ago
        separate tradeoff worth naming - do you want "memories" available within session vs after the conversation has ended? that was what i was trying to convey in this paragraph
  • chrismsimpson 4 minutes ago
    A seminal post
    • brgsk 2 minutes ago
      thanks for reading
  • cpard 1 hour ago
    This is a great post and I really appreciate making the cognitive science terminology clear.

    the author is doing a great job telling what is missing from the current memory frameworks for agents but what is missing in my opinion is also an argument about the necessity or not of these missing components.

    • brgsk 59 minutes ago
      fair — this post mapped the gaps without making the case for whether filling them changes what an agent can do. the interesting ones are procedural and prospective. both deserve their own post.

      thanks for the read.

      • cpard 42 minutes ago
        Hopefully I didn’t sound too critical of the post because this wasn’t my intention. The post delivered what was needed and thank you for this!

        The reason I asked the question is because in the case we don’t need the rest, it would be better to not use this terminology for these systems. We already anthropomorphize LLMs too much and although I get the marketing value of that, it’s not always to the benefit of the people who interact with them.

        Please do write the rest of the posts!

        • brgsk 34 minutes ago
          not at all, I appreciate your comments!

          yeah i agree with you on not using the terminology, although it's intuitive it's also confusing enough. it's tempting to do that, but i share your sentiment

      • behat 43 minutes ago
        Thanks for writing this, and look forward to the one on procedural memory.

        Seems like teams are encoding procedural knowledge in skills repositories, and I wonder if there’s additional utility from an auto created procedural memory layer

        • brgsk 27 minutes ago
          [flagged]
  • joemoon 48 minutes ago
    Spidey senses going off here. The first two comments read like an LLM.
    • brgsk 45 minutes ago
      yeah i used cc to help me write the post itself and the comment, my bad
      • arkmm 38 minutes ago
        everything is computer
        • brgsk 32 minutes ago
          thats beautiful! wow!
  • shenli3514 39 minutes ago
    [flagged]
  • markhuang 34 minutes ago
    [dead]