跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.20230 · v1

A Matrix with Steps: An Episodic and Procedural Memory Method for Tool-Using Agents and Robots

首次提交 2026-11-16 04:06
版本更新 2026-11-16 04:06
雷达观测 2026-11-17 12:06
arXiv ↗PDF ↗

摘要

Agents that call tools and robots that execute tasks are challenged by two shortcomings: they repeat what failed before, and they learn from runs that they only believe succeeded. We describe the construction of a memory that addresses both, built for a production agent platform. Every run is recorded as an episode, and only an outcome confirmed by an external check, a test, a tool result or a person counts; the agent's own claim never does. What is learned from episodes is kept in a memory with two parts: an associative matrix, written with the delta rule, which finds the procedure that applies to a situation; and a section of ordered, addressable step records that answers what comes after a given step, where a step is a goal of the agent, not a motion. Because the matrix sums its associations, it keeps no order and cannot return or delete a single one; it serves as the finder, and order, outcome counts and deletion live in the records. We describe how execution outcomes are stored, how episodes are recalled and used for solving tasks, and the file format that stores the memory. In simulation, a robot that judged its own success and stopped when it believed it was done reported 96%96\% of its tasks as done and solved 35%35\%; with external verification it solved $79%$. In a store of 100000100000 synthetic episodes (457457MB on disk), the oldest episode was still found first, no episode from another person's memory was returned, and a success vouched for only by the model was never recalled; a lookup took 1616ms on a laptop and 4040ms on a server node (median). Task outcomes were measured up to about 1000010000 confirmed records, where the exact records solved every revisit at the first attempt.

作者备注:17 pages, 5 figures

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-16 04:06 · 最新观测版本

A Matrix with Steps: An Episodic and Procedural Memory Method for Tool-Using Agents and Robots

Agents that call tools and robots that execute tasks are challenged by two shortcomings: they repeat what failed before, and they learn from runs that they only believe succeeded. We describe the construction of a memory that addresses both, built for a production agent platform. Every run is recorded as an episode, and only an outcome confirmed by an external check, a test, a tool result or a person counts; the agent's own claim never does. What is learned from episodes is kept in a memory with two parts: an associative matrix, written with the delta rule, which finds the procedure that applies to a situation; and a section of ordered, addressable step records that answers what comes after a given step, where a step is a goal of the agent, not a motion. Because the matrix sums its associations, it keeps no order and cannot return or delete a single one; it serves as the finder, and order, outcome counts and deletion live in the records. We describe how execution outcomes are stored, how episodes are recalled and used for solving tasks, and the file format that stores the memory. In simulation, a robot that judged its own success and stopped when it believed it was done reported 96%96\% of its tasks as done and solved 35%35\%; with external verification it solved $79%$. In a store of 100000100000 synthetic episodes (457457MB on disk), the oldest episode was still found first, no episode from another person's memory was returned, and a success vouched for only by the model was never recalled; a lookup took 1616ms on a laptop and 4040ms on a server node (median). Task outcomes were measured up to about 1000010000 confirmed records, where the exact records solved every revisit at the first attempt.

查看此版本解读 →查看此版本原文 ↗