惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Engineering at Meta
Engineering at Meta
T
Threatpost
P
Palo Alto Networks Blog
NISL@THU
NISL@THU
O
OpenAI News
Project Zero
Project Zero
G
GRAHAM CLULEY
P
Privacy International News Feed
A
Arctic Wolf
Microsoft Azure Blog
Microsoft Azure Blog
H
Help Net Security
M
MIT News - Artificial intelligence
T
Threat Research - Cisco Blogs
S
Security @ Cisco Blogs
Google DeepMind News
Google DeepMind News
B
Blog RSS Feed
D
Docker
aimingoo的专栏
aimingoo的专栏
博客园 - 【当耐特】
N
Netflix TechBlog - Medium
云风的 BLOG
云风的 BLOG
雷峰网
雷峰网
W
WeLiveSecurity
P
Proofpoint News Feed
腾讯CDC
Cloudbric
Cloudbric
S
Secure Thoughts
C
Check Point Blog
博客园 - Franky
T
The Exploit Database - CXSecurity.com
T
Troy Hunt's Blog
GbyAI
GbyAI
Security Archives - TechRepublic
Security Archives - TechRepublic
Application and Cybersecurity Blog
Application and Cybersecurity Blog
月光博客
月光博客
C
Cyber Attacks, Cyber Crime and Cyber Security
I
Intezer
TaoSecurity Blog
TaoSecurity Blog
L
Lohrmann on Cybersecurity
V
Visual Studio Blog
F
Fortinet All Blogs
博客园 - 叶小钗
C
CXSECURITY Database RSS Feed - CXSecurity.com
CTFtime.org: upcoming CTF events
CTFtime.org: upcoming CTF events
Recorded Future
Recorded Future
C
Cisco Blogs
博客园 - 司徒正美
Stack Overflow Blog
Stack Overflow Blog
Y
Y Combinator Blog
Apple Machine Learning Research
Apple Machine Learning Research

博客园 - fariver

[PaperWritting] 多模态大模型架构图摘录 每日Paper - 2026-03-06 每日Paper | 2026年3月4日 [Paper Reading] Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking [PaperReading] OneSearch A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search [PaperReading] OneRec Technical Report [Paper Reading] Tiger: Recommender Systems with Generative Retrieval [PaperReading] Generative Recommendation with Semantic IDs: A Practitioner’s Handbook [PaperReading] GME: Improving Universal Multimodal Retrieval by Multimodal LLMs [Paper Reading] UniME-V2: MLLM-as-a-Judge for Universal Multimodal Embedding Learning [PaperReading] UniME: Breaking the Modality Barrier: Universal Embedding Learning with Multimodal LLMs [PaperReading] Qwen2.5-VL Technical Report [PaperReading] DeepSeek-OCR: Contexts Optical Compression [PaperReading] SAIL-Embedding Technical Report: Omni-modal Embedding Foundation Model [PaperReading] VLM2Vec-V2: Advancing Multimodal Embedding for Videos, Images, and Visual Documents [PaperReading] VLM2VEC: TRAINING VISION-LANGUAGE MODELS FOR MASSIVE MULTIMODAL EMBEDDING TASKS [PaperReading] REACT: SYNERGIZING REASONING AND ACTING IN LANGUAGE MODELS [PaperReading] Qwen2-VL: Enhancing Vision-Language Model’s Perception of the World at Any Resolution [PaperReading] Mind Search: Mimicking Human Minds Elicits Deep AI Searcher [PaperReading] METAGPT: META PROGRAMMING FOR A MULTI-AGENT COLLABORATIVE FRAMEWORK
[PaperReading] MemGPT: Towards LLMs as Operating Systems
fariver · 2025-09-23 · via 博客园 - fariver

MemGPT: Towards LLMs as Operating Systems

link
时间:23.10
单位:UC Berkeley
相关领域:Agent
作者相关工作:Charles Packer
被引次数:238
项目主页:
https://research.memgpt.ai/

TL;DR

大语言模型上下文长度限制了用户多轮对话以及文档分析的需求,本文参考传统计算机操作系统中多级内存管理机制,提出MemGPT算法,能够智能地管理LLM存储的多个层级。实验证明该方法能够支持long-term交互。

Method

image

Main context

  • system instructions: 保存MemGPT control flow信息,例如,如何使用不同层级内存,使用MemGPT functions的指标 等。
  • working context: unstructured text,用来存储一些关键信息,例如,用户画像,用户偏好 等。
  • FIFO Queue: 存储历史滚动信息,第一个Key是历史所有信息的概括。
    FIFO Queue:
  1. 消息生命周期管理
  • ​​接收与写入​​
  • 持久化存储: 所有会话信息将永久性地写入外部的“回忆存储”(Recall Storage)。
  • 检索与重载​​:当LLM通过函数调用(如搜索)从回忆存储中检索历史消息时,这些消息会被重新载入FIFO队列中。
  1. 上下文溢出控制
  • 警告:队列填充达到70%时,警告MemGPT及时将重要工作归档。
  • 驱逐:队列填充达到70%时,执行强制性的​​队列刷新(Flushing)​​ 操作,从队列​​头部​​(最旧的消息)开始​​驱逐一定数量的消息​​。
  • 递归摘要生成:从现有的递归摘要​​(存储在FIFO队列的第一个索引中)和​​本次被驱逐的消息​​,生成一个​​新的、更精简的递归摘要​​。

Q:图3中的Archival Storage与Recall Storage有什么区别?

  • ​​Archival Storage​​ 像一个​​主动搜集的研究资料库​​,用于存储LLM​​主动选择​​的、与任务相关的原始信息片段。
  • Recall Storage​​ 像一个​​被动记录的完整对话日志​​,用于​​自动记录​​所有发生过的交互的完整历史。

Experiment

评估的考查维度:

image
image

总结与思考

参考操作系统内存管理设计,但整体框架依然偏策略,不知道后续有没有 基于RL更新Agent中LLM 或者 端到端Agent的方法。

相关链接

cool paper链接:https://papers.cool/
https://zhuanlan.zhihu.com/p/711289611