7 comments

  • swyx20 minutes ago
    hmm the repo doesnt mention this at all but this name and problem domain brings up HippoRAG <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;2405.14831" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;2405.14831</a> &lt;- any relation? seems odd to miss out this exactly similarly named paper with related techniques.
  • kami2328 minutes ago
    Cool to see others on this thread.<p>Here&#x27;s a post I wrote about how we can start to potentially mimic mechanisms<p><a href="https:&#x2F;&#x2F;n0tls.com&#x2F;2026-03-14-musings.html" rel="nofollow">https:&#x2F;&#x2F;n0tls.com&#x2F;2026-03-14-musings.html</a><p>Would love to compare notes, I&#x27;m also looking at linguistic phenomena through an LLM lens<p><a href="https:&#x2F;&#x2F;n0tls.com&#x2F;2026-03-19-more-musings.html" rel="nofollow">https:&#x2F;&#x2F;n0tls.com&#x2F;2026-03-19-more-musings.html</a><p>Hoping to wrap up some of the kaggle eval work and move back to researching more neuropsych.
  • nberkman1 hour ago
    Cool project. I like the neuroscience analogy with decay and consolidation.<p>I&#x27;ve been working on a related problem from the other direction: Claude Code and Codex already persist full session transcripts, but there&#x27;s no good way to search across them. So I built ccrider (<a href="https:&#x2F;&#x2F;github.com&#x2F;neilberkman&#x2F;ccrider" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;neilberkman&#x2F;ccrider</a>). It indexes existing sessions into SQLite FTS5 and exposes an MCP server so agents can query their own conversation history without a separate memory layer. Basically treating it as a retrieval problem rather than a storage problem.
  • the_arun1 hour ago
    Aren&#x27;t tools like claude already store context by project in file system? Also any reason use &quot;capture&quot; instead of &quot;export&quot; (an obvious opposite of import)?
    • nberkman31 minutes ago
      &gt; Aren&#x27;t tools like claude already store context by project in file system?<p>They do, the missing piece is a tool to access them. See comment about my tool that addresses this: <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=47668270">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=47668270</a>
  • gfody37 minutes ago
    yegge has a cool solution for this in gastown: the current agent is able to hold a seance with the previous one
  • esafak17 minutes ago
    How does it select what to forget? Let&#x27;s say I land a PR that introduces a sharp change, migrating from one thing to another. An exponential decay won&#x27;t catch this. Biological learning makes sense when things we observe similar things repeatedly in order to learn patterns. I am skeptical that it applies to learning the commits of one code base.
  • cyanydeez1 hour ago
    no open code plugin? This seems like something that should just run in the background. It&#x27;s well documented that it should just be a skill agents can use when they get into various fruitless states.<p>The &quot;biological&quot; memory strength shouldn&#x27;t just be a time thing, and even then, the time of the AI agent should only be conformed to the AI&#x27;s lifetime and not the actual clock. Look up <a href="https:&#x2F;&#x2F;stackoverflow.com&#x2F;questions&#x2F;3523442&#x2F;difference-between-clock-realtime-and-clock-monotonic" rel="nofollow">https:&#x2F;&#x2F;stackoverflow.com&#x2F;questions&#x2F;3523442&#x2F;difference-betwe...</a> monotonic clock. If you want a decay, it shouldn&#x27;t be related to an actual clock, but it&#x27;s work time.<p>But memory is more about triggers than it is about anything else. So you should absolutely have memory triggers based on location. Something like a path hash. So whever an agent is working and remembering things it should be tightly compacted to that location; only where a &quot;compaction&quot; happens should these memories become more and more generalized to locations.<p>The types of memory that often are more prominent are like this, whether it&#x27;s sports or GUIs, physical location triggers much more intrinsics than conscious memory. Focus on how to trigger recall based on project paths, filenames in the path, file path names, etc.
    • pbhjpbhj19 minutes ago
      Memory links to location but that&#x27;s largely because humans are localised. Isn&#x27;t that also a weakness. We should be trying to exploit the benefits of non-locality [of ML models and training data] too.<p>I feel like much of my life is virtual, non-localised. Writing missives to the four corners of the wind here and elsewhere; gaming online; research&#x2F;chats with LLMs or on the web, email with people.<p>My physical location is often not important - a continuing context from non-physical aspects of my existence matters more.<p>That said, one of the things that&#x27;s hard for me about digital life is the lack of waymarks - I used to be quite &quot;geographical&quot; in my thinking. Like &quot;oh the part I found interesting was on the left page after the RGB diagram&quot;, I&#x27;d find that and also find my train of thought and extend it. Now, information can be in any myriad of freeform places across at least 3 devices and in emails, notebooks, bookmarks, chat histories, and of course my brain. When some ready syncretism of those things happens it feels like we&#x27;ll make better advances. Personal agents can be a part of that.
    • Grosvenor1 hour ago
      Sparse distributed memory is what you’re looking for.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Sparse_distributed_memory" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Sparse_distributed_memory</a>
    • kitfunso1 hour ago
      coming right up, adding it as we speak
    • russellthehippo1 hour ago
      yep came here to say this. great to hear it&#x27;s in process.