What question did this study set out to answer?

This research aims to address the limitations of standard retrieval-augmented generation in large language models by introducing a novel cognitive memory architecture.

May 31, 2026Open Access

Astrum Verum: Compositional Episodic Cognitive Memory for LLM Agents via Vector Symbolic Architectures

Key Points

This research aims to address the limitations of standard retrieval-augmented generation in large language models by introducing a novel cognitive memory architecture.
Developed a composition-episodic cognitive memory architecture using vector symbolic architectures.
Evaluated capacity scaling, episodic order recall, and semantic fidelity in LLMs with a high-dimensional memory space.
Compared the performance of the new architecture against standard retrieval-augmented generation baselines.
Achieved graceful capacity scaling up to 16,000 facts.
Demonstrated near-perfect episodic order recall within a bounded working window.
Outperformed standard RAG on role-ambiguous queries with maintained semantic fidelity.

Abstract

Large Language Models (LLMs) equipped with standard Retrieval-Augmented Generation (RAG) suffer from structural blindness: they cannot distinguish between role-swapped propositions due to the commutative nature of continuous vector search. We introduce a composition-episodic cognitive memory architecture based on Vector Symbolic Architectures (VSA). By decoupling semantic embedding from structural role-binding, the system preserves exact structural relationships and sequential episodic order without sacrificing continuous generalization. We demonstrate that the architecture sustains graceful capacity scaling up to 16,000 facts, achieves near-perfect episodic order recall within a bounded working window, and maintains high semantic fidelity between the original dense embedding space and the high-dimensional space. Our results show that this hybrid neuro-symbolic approach significantly outperforms standard RAG baselines on role-ambiguous queries while providing a mathematically rigorous foundation for verifiable AI memory.

Read Full Paperexternally

Mark Helpful

Bookmark

Relay

View Full Paper