Memory Sandbox: Transparent and Interactive Memory Management for Conversational Agents

TL;DR

Memory Sandbox is an interactive system that enables users to view, control, and manage the memory of conversational agents powered by large language models, improving transparency and user control.

Contribution

It introduces a novel interactive system that allows users to manage conversational memory explicitly, addressing the lack of user affordances in existing LLM-based agents.

Findings

Users can view and manipulate conversation memories.

Memory management improves user understanding of the agent.

The system supports sharing and summarizing memories.

Abstract

The recent advent of large language models (LLM) has resulted in high-performing conversational agents such as chatGPT. These agents must remember key information from an ongoing conversation to provide responses that are contextually relevant to the user. However, these agents have limited memory and can be distracted by irrelevant parts of the conversation. While many strategies exist to manage conversational memory, users currently lack affordances for viewing and controlling what the agent remembers, resulting in a poor mental model and conversational breakdowns. In this paper, we present Memory Sandbox, an interactive system and design probe that allows users to manage the conversational memory of LLM-powered agents. By treating memories as data objects that can be viewed, manipulated, recorded, summarized, and shared across conversations, Memory Sandbox provides interaction affordances for users to manage how the agent should `see' the conversation.