Jev-Mem:受 System-One/System-Two 启发的智能体记忆架构

elvis · @omarsar0 · X·2026-09-26 05:51·37分钟前
AI 导读

Jev-Mem 是一种受 System-One/System-Two 认知启发的新型智能体记忆架构,将记忆构建提速 6.6 倍、查询延迟降低 36.7%。在 LoCoMo 上,它以 LLM 评审 0.777 的总分较最强基线相对提升 11.0%,记忆构建耗时 158 秒,平均查询延迟降至 0.93 秒。

elvis@omarsar0
45AI 编辑部评分,满分 100

Jev-Mem:受 System-One/System-Two 启发的智能体记忆架构

2026-09-26 05:51· 37分钟前
AI 导读

Jev-Mem 是一种受 System-One/System-Two 认知启发的新型智能体记忆架构,将记忆构建提速 6.6 倍、查询延迟降低 36.7%。在 LoCoMo 上,它以 LLM 评审 0.777 的总分较最强基线相对提升 11.0%,记忆构建耗时 158 秒,平均查询延迟降至 0.93 秒。

Banger report on building faster memory for AI Agents.

If your agent's memory layer is slow, this design is worth a look.

Great weekend read.

It proposes a new agentic memory architecture, Jev-Mem, inspired by System-One/System-Two cognition.

This builds agent memory 6.6x faster and cuts query latency by 36.7% while scoring higher on LoCoMo.

In this system, a lightweight controller makes the memory decisions.

It assigns memory types and relations when storing, and handles routing, retrieval budget, graph traversal, scoring and stopping when reading. The LLM is called only for final reasoning and answer writing.

On LoCoMo, it scores 0.777 overall with an LLM judge, an 11.0% relative improvement over the strongest baseline. Memory construction takes 158 seconds, and average query latency drops to 0.93 seconds.

Paper: https://academy.dair.ai/papers/jev-mem-system-one-controlled-agentic-memory-for-efficient-ai-agents-2609.23986

来源:elvis· x.com