Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Suli, Duan, Yiqun, Deng, Yu, Zhao, Rundong, Shi, Dai, Zhou, Xinliang
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.13438
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918520729632768
author Wang, Suli
Duan, Yiqun
Deng, Yu
Zhao, Rundong
Shi, Dai
Zhou, Xinliang
author_facet Wang, Suli
Duan, Yiqun
Deng, Yu
Zhao, Rundong
Shi, Dai
Zhou, Xinliang
contents Existing agent memory remains predominantly reactive and retrieval-based, lacking the capacity to autonomously organize experience into persistent cognitive structure. Toward genuinely autonomous agents, we introduce CogniFold, a brain-inspired "always-on" agent memory designed for the next generation of proactive assistants. CogniFold continuously folds fragmented event streams into self-emerging cognitive structures, bootstrapping progressively higher-level cognition from incoming events and accumulated knowledge. We ground this by extending Complementary Learning Systems (CLS) theory from two layers (hippocampus, neocortex) to three, adding a prefrontal intent layer. Emulating the prefrontal cortex as the locus of intentional control and decision-making, CogniFold achieves this through graph-topology self-organization: cognitive structures proactively assemble under the stream, merge when semantically similar, decay when stale, relink through associative recall, and surface intents when concept-cluster density crosses a threshold. We evaluate structural formation using CogEval-Bench, demonstrating that CogniFold uniquely produces memory structures that match cognitive expectations and concept emergence. Furthermore, across 7 broad-coverage benchmarks spanning five cognitive domains, we validate that CogniFold simultaneously performs robustly on conventional memory benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2605_13438
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CogniFold: Always-On Proactive Memory via Cognitive Folding
Wang, Suli
Duan, Yiqun
Deng, Yu
Zhao, Rundong
Shi, Dai
Zhou, Xinliang
Artificial Intelligence
Computation and Language
Existing agent memory remains predominantly reactive and retrieval-based, lacking the capacity to autonomously organize experience into persistent cognitive structure. Toward genuinely autonomous agents, we introduce CogniFold, a brain-inspired "always-on" agent memory designed for the next generation of proactive assistants. CogniFold continuously folds fragmented event streams into self-emerging cognitive structures, bootstrapping progressively higher-level cognition from incoming events and accumulated knowledge. We ground this by extending Complementary Learning Systems (CLS) theory from two layers (hippocampus, neocortex) to three, adding a prefrontal intent layer. Emulating the prefrontal cortex as the locus of intentional control and decision-making, CogniFold achieves this through graph-topology self-organization: cognitive structures proactively assemble under the stream, merge when semantically similar, decay when stale, relink through associative recall, and surface intents when concept-cluster density crosses a threshold. We evaluate structural formation using CogEval-Bench, demonstrating that CogniFold uniquely produces memory structures that match cognitive expectations and concept emergence. Furthermore, across 7 broad-coverage benchmarks spanning five cognitive domains, we validate that CogniFold simultaneously performs robustly on conventional memory benchmarks.
title CogniFold: Always-On Proactive Memory via Cognitive Folding
topic Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2605.13438