Cognitive Workspace: Active Memory Management for LLMs An Empirical Study of Functional Infinite Context
Tao An Hawaii Pacific University https://github.com/tao-hpu
Abstract
Large Language Models (LLMs) face fundamental limitations in context management despite recent advances extending context windows to millions of tokens. We propose Cognitive Workspace, a novel paradigm that transcends traditional Retrieval-Augmented Generation (RAG) by emulating human cognitive mechanisms of external memory use. Drawing from cognitive science foundations including Baddeley’s working memory model [1, 2], Clark’s extended mind thesis [3], and Hutchins’ distributed cognition framework [4], we demonstrate that current passive retrieval systems fail to capture the dynamic, task-driven nature of human memory management. Our analysis of 2024-2025 developments reveals that while techniques like Infini-attention [5] and StreamingLLM [6] achieve impressive context lengths, they lack the metacognitive awareness and active planning capabilities essential for true cognitive extension. Cognitive Workspace addresses these limitations through three core innovations: (1) active memory management with deliberate information curation, (2) hierarchical cognitive buffers enabling persistent working states, and (3) task-driven context optimization that dynamically adapts to cognitive de
中文速览
大语言模型（LLM）在处理超长文本时，既记不住关键信息、又无法像人一样主动管理记忆——现有的检索增强生成（RAG）方法只是被动地按需查找，本质上仍是"外挂数据库"而非真正的认知系统。受巴德利工作记忆模型、克拉克延展心智理论等认知科学成果启发，作者提出"认知工作区"（Cognitive Workspace）框架，通过主动记忆管理、分层认知缓冲区和任务驱动的上下文优化三大机制，让模型像人用白板一样动态整理、保留和调用信息。实验验证显示，认知工作区的记忆复用率平均达58.6%，而传统RAG为0%，综合效率净提升17–18%，统计显著性极高（p<0.001，Cohen's d>23）。这项工作为LLM真正实现"无限上下文"提供了一条有别于单纯扩大窗口的新路径，将记忆从被动检索升级为认知系统的有机组成部分。
原文 arXiv:2508.13171;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2508.13171v1