A Multi-Memory Segment System for Generating High-Quality Long-Term Memory Content in Agents

TL;DR

This paper introduces a multi-memory segment system inspired by cognitive psychology to generate high-quality long-term memory content in agents, improving retrieval and response quality.

Contribution

The proposed MMS system processes short-term memory into multiple segments and constructs specialized memory units, enhancing long-term memory content quality in agents.

Findings

Effective in improving recall performance and response quality.

Demonstrated robustness with varying input memory sizes.

Validated through experiments on the LoCoMo dataset.

Abstract

In the current field of agent memory, extensive explorations have been conducted in the area of memory retrieval, yet few studies have focused on exploring the memory content. Most research simply stores summarized versions of historical dialogues, as exemplified by methods like A-MEM and MemoryBank. However, when humans form long-term memories, the process involves multi-dimensional and multi-component generation, rather than merely creating simple summaries. The low-quality memory content generated by existing methods can adversely affect recall performance and response quality. In order to better construct high-quality long-term memory content, we have designed a multi-memory segment system (MMS) inspired by cognitive psychology theory. The system processes short-term memory into multiple long-term memory segments, and constructs retrieval memory units and contextual memory units based on these segments, with a one-to-one correspondence between the two. During the retrieval phase, MMS will match the most relevant retrieval memory units based on the user's query. Then, the corresponding contextual memory units is obtained as the context for the response stage to enhance knowledge, thereby effectively utilizing historical data. We conducted experiments on the LoCoMo dataset and further performed ablation experiments, experiments on the robustness regarding the number of input memories, and overhead experiments, which demonstrated the effectiveness and practical value of our method.