# Multi-Layered Memory Architectures for LLM Agents: An Experimental Evaluation of Long-Term Context Retention

arXiv: https://arxiv.org/abs/2603.29194 · HTML: https://mlsanity.com/paper/2603.29194/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Agent memory systems, one of 16 directions in LLM agents and tool use.
- Builds on 8 of its 22 references that are in the map.
- First posted on arXiv 31 Mar 2026.

## Placed in

- LLM agents and tool use / [Agent memory systems](https://mlsanity.com/agents/agent-memory-systems/)

## Builds on (most cited first, 8 of 8)

- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory](https://mlsanity.com/paper/2402.04617/) (2024)
- [Memory OS of AI Agent](https://mlsanity.com/paper/2506.06326/) (2025)
- [HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model](https://mlsanity.com/paper/2408.09559/) (2024)
- [LAVa: Layer-wise KV Cache Eviction with Dynamic Budget Allocation](https://mlsanity.com/paper/2509.09754/) (2025)
- [In-Context Former: Lightning-fast Compressing Context for Large Language Model](https://mlsanity.com/paper/2406.13618/) (2024)
- [Interpersonal Memory Matters: A New Task for Proactive Dialogue Utilizing Conversational History](https://mlsanity.com/paper/2503.05150/) (2025)
- [ATACompressor: Adaptive Task-Aware Compression for Efficient Long-Context Processing in LLMs](https://mlsanity.com/paper/2602.03226/) (2026)

## Built on by (most cited first, 2 of 2)

- [ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture for Autonomous AI Systems](https://mlsanity.com/paper/2604.23878/) (2026)
- [CreaMem: A Scene-Aware Memory Architecture for Personalized Agents](https://mlsanity.com/paper/2609.08550/) (2026)

## Abstract (arXiv)

Long-horizon dialogue systems suffer from semanticdrift and unstable memory retention across extended sessions. This paper presents a Multi-Layer Memory Framework that decomposes dialogue history into working, episodic, and semantic layers with adaptive retrieval gating and retention regularization. The architecture controls cross-session drift while maintaining bounded context growth and computational efficiency. Experiments on LOCOMO, LOCCO, and LoCoMo show improved performance, achieving 46.85 Success Rate, 0.618 overall F1 with 0.594 multi-hop F1, and 56.90% six-period retention while reducing false memory rate to 5.1% and context usage to 58.40%. Results confirm enhanced long-term retention and reasoning stability under constrained context budgets.