CraniMem: Cranial Inspired Gated and Bounded Memory for Agentic Systems

TL;DR

CraniMem introduces a neurocognitively inspired, gated, and bounded memory system for LLM agents, enhancing long-term retention, robustness, and interference management in long-running workflows.

Contribution

The paper proposes CraniMem, a novel memory architecture combining goal-conditioned gating, utility tagging, episodic buffers, and knowledge graphs for improved agent memory management.

Findings

Outperforms Vanilla RAG and Mem0 in robustness tests.

Maintains better performance under noisy and distracting inputs.

Effectively balances memory growth and interference through scheduled consolidation.

Abstract

Large language model (LLM) agents are increasingly deployed in long running workflows, where they must preserve user and task state across many turns. Many existing agent memory systems behave like external databases with ad hoc read/write rules, which can yield unstable retention, limited consolidation, and vulnerability to distractor content. We present CraniMem, a neurocognitively motivated, gated and bounded multi-stage memory design for agentic systems. CraniMem couples goal conditioned gating and utility tagging with a bounded episodic buffer for near term continuity and a structured long-term knowledge graph for durable semantic recall. A scheduled consolidation loop replays high utility traces into the graph while pruning low utility items, keeping memory growth in check and reducing interference. On long horizon benchmarks evaluated under both clean inputs and injected noise, CraniMem is more robust than a Vanilla RAG and Mem0 baseline and exhibits smaller performance drops under distraction. Our code is available at https://github.com/PearlMody05/Cranimem and the accompanying PyPI package at https://pypi.org/project/cranimem.