A Scenario-Driven Cognitive Approach to Next-Generation AI Memory

2025/09/16 by Cai, Linyue, Yuyang Cheng, Cheng, Yuyang +9 · 2 citations
Computer Science · Engineering · #AI-based Problem Solving and Planning #Artificial Intelligence (cs.AI) #Cognitive Computing and Networks #FOS: Computer and information sciences #Robotics and Automated Systems

paper · pdf · doi:10.48550/arxiv.2509.13235

openalex publication_date 2025/09/16 · openalex created_date 2025/10/18 · openalex updated_date 2026/07/28

Abstract

As artificial intelligence advances toward artificial general intelligence (AGI), the need for robust and human-like memory systems has become increasingly evident. Current memory architectures often suffer from limited adaptability, insufficient multimodal integration, and an inability to support continuous learning. To address these limitations, we propose a scenario-driven methodology that extracts essential functional requirements from representative cognitive scenarios, leading to a unified set of design principles for next-generation AI memory systems. Based on this approach, we introduce the COgnitive Layered Memory Architecture (COLMA), a novel framework that integrates cognitive scenarios, memory processes, and storage mechanisms into a cohesive design. COLMA provides a structured foundation for developing AI systems capable of lifelong learning and human-like reasoning, thereby contributing to the pragmatic development of AGI.

Citations

• Memp: Exploring Agent Procedural Memory
• MIRIX: Multi-Agent Memory System for LLM-Based Agents
• HAWK: A Hierarchical Workflow Framework for Multi-Agent Collaboration
• MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents
• Rethinking Memory in LLM based Agents: Representations, Operations, and Emerging Topics
• Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
• A-MEM: Agentic Memory for LLM Agents
• Key-value memory in the brain
• Agent AI with LangGraph: A Modular Framework for Enhancing Machine Translation Using Large Language Models
• Exploration of LLM Multi-Agent Application Implementation Based on LangGraph+CrewAI
• Retrieval-Augmented Generation for Large Language Models: A Survey
• A Survey of Large Language Models
• CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
• Self-recovery of memory via generative replay
• Relating transformers to models and neural representations of the hippocampal formation
• Knowledge Graphs
• Experience Replay for Continual Learning
• Relational inductive biases, deep learning, and graph networks
• Overcoming catastrophic forgetting with hard attention to the task
• FearNet: Brain-Inspired Model for Incremental Learning
• Multimodal Machine Learning: A Survey and Taxonomy
• Multimodal Machine Learning: A Survey and Taxonomy
• Modeling Relational Data with Graph Convolutional Networks
• Billion-scale similarity search with GPUs
• Towards A Rigorous Science of Interpretable Machine Learning
• Memory: Organization and Control
• Hippocampal sharp wave‐ripple: A cognitive biomarker for episodic memory and planning
• Challenges in Representation Learning: A report on three machine\n learning contests
• Representation Learning: A Review and New Perspectives
• Representation Learning: A Review and New Perspectives
• Catastrophic Forgetting, Rehearsal and Pseudorehearsal
• An Integrative Theory of Prefrontal Cortex Function

Cited by

• Memory Bear AI A Breakthrough from Memory to Cognition Toward Artificial General Intelligence
• Mem-α: Learning Memory Construction via Reinforcement Learning

Related

• CAIM: Development and Evaluation of a Cognitive AI Memory Framework for Long-Term Interaction with Intelligent Agents
• Contextual Memory Intelligence -- A Foundational Paradigm for Human-AI Collaboration and Reflective Generative AI Systems
• AI Meets Brain: Memory Systems from Cognitive Neuroscience to Autonomous Agents
• Human-inspired Perspectives: A Survey on AI Long-term Memory
• Cognitive AI framework: advances in the simulation of human thought