Memora: A Harmonic Memory Representation Balancing Abstraction and Specificity

2026-02-03

AI Summary

This work addresses the challenge that existing agent memory systems struggle to simultaneously maintain abstraction and detail specificity when scaled, leading to degraded retrieval efficiency and reasoning performance. To resolve this, we propose Harmonic Memory Representation (HMR), a novel architecture that structurally balances these two aspects within a unified framework. HMR employs a primary abstract index to anchor and dynamically integrate specific memories, while leveraging cue-based anchors to construct multi-dimensional retrieval pathways that enable context-aware, efficient access. Theoretically, we show that conventional RAG and knowledge graph–based memory systems are special cases of our formulation. Empirical evaluations on the LoCoMo and LongMemEval benchmarks establish new state-of-the-art results, demonstrating significant improvements in retrieval relevance and reasoning capability under large-scale memory settings.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Knowledge Representation and Reasoning: Argumentation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Agent memory systems must accommodate continuously growing information while supporting efficient, context-aware retrieval for downstream tasks. Abstraction is essential for scaling agent memory, yet it often comes at the cost of specificity, obscuring the fine-grained details required for effective reasoning. We introduce Memora, a harmonic memory representation that structurally balances abstraction and specificity. Memora organizes information via its primary abstractions that index concrete memory values and consolidate related updates into unified memory entries, while cue anchors expand retrieval access across diverse aspects of the memory and connect related memories. Building on this structure, we employ a retrieval policy that actively exploits these memory connections to retrieve relevant information beyond direct semantic similarity. Theoretically, we show that standard Retrieval-Augmented Generation (RAG) and Knowledge Graph (KG)-based memory systems emerge as special cases of our framework. Empirically, Memora establishes a new state-of-the-art on the LoCoMo and LongMemEval benchmarks, demonstrating better retrieval relevance and reasoning effectiveness as memory scales.

Problem

• agent memory
• abstraction
• specificity
• memory representation
• context-aware retrieval

Innovation

• harmonic memory representation
• abstraction-specificity balance
• cue anchors
• memory retrieval policy
• Retrieval-Augmented Generation