Skip to content

Latest commit

 

History

1 Commit

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 

Repository files navigation

Stream Memory

Agent memory research code: a third path between "absorb everything into the weights" and "store everything in an external retrieval index."

Motivation

Most agent-memory designs assume one of two extremes: consolidate experience into model weights (slow, destructive, opaque) or keep an ever-growing external store and retrieve from it (cheap to write, hard to keep predictive). This project studies an old and under-used alternative from the online-learning literature: maintain several local predictive models simultaneously, update all of them online as the stream arrives, and at each decision point act on whichever currently predicts best. In a world too big to model globally, selection over cheap local models can beat both extremes — and the experiments here measure when and why, including a full evaluation on MemoryAgentBench.

What is in this repository

  • artifact/ — the frozen reproducibility package, preserved byte for byte:
    • code/ — every generator and analyser behind the reported numbers (stream construction, matched-budget baselines, non-stationarity and abstention experiments, follow-the-leading-history traces, agent evaluation, figure generators)
    • CLAIM-*.md — analysis plans written before the runs and shipped unchanged
    • CORRECTIONS.md — every correction and retraction made during the work
    • MANIFEST.txt + verify.sh — SHA-256 of every shipped file and a checker

Reproducing

cd artifact
sh verify.sh          # confirms every file matches MANIFEST.txt
cat README.md         # per-experiment entry points and environment notes

Integrity note

The artifact/ tree is exactly the package the reported numbers were generated from; it is intentionally not edited here (verify with sh verify.sh). Anything at the repository root is packaging only.

Status

Research code release. A paper reference will be added here once available.

License

MIT (see LICENSE).

About

Online model-selection memory for agents: maintain many local predictors, update all, act on the best

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages