Boy I'd like to read a compact non-LLM version of the key concepts here. The signal ratio is very low, and crafted with weird LLM-isms throughout, and very hard to parse.

I've been experimenting with mapping a zettelkasten system over to my agents with a few goals in mind, not least applying the idea of more 'test time compute' to the storing of memories as a way to add useful structure that can be tapped later during retrieval. (github.com/vessenes/zet - MIT license - no warranties)

There's some good and some bad, but I think it's better than just a raw embedding memory store for agents. It's definitely better for a human in that it's navigable and understandable, while remaining useful for agents.

But, I'd really like to read more about the space and get ideas -- this blog post was just too difficult to parse for me, sadly.

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?

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.

It seems most memory systems are built ETL style, when they'd be better off ELT style. As the core memory systems and LLMs improve, you'd be better off querying from raw source information rather than whatever consolidation of that information was deemed appropriate at the time. That is, just store it all compressed, then build (or rebuild) the memories as new information enters or as architecture evolves

  multimodal models with environmental grounding may eventually have an analog [of affect]. text-only agents can’t.

I'd be curious to hear an explanation about why text can't. Or equivalently, why multimodal could.

Spidey senses going off here. The first two comments read like an LLM.

A seminal post