Crafting Personalized Agents through Retrieval-Augmented Generation on Editable Memory Graphs
This addresses the challenge of personalizing AI assistants for users by leveraging their smartphone data, representing an incremental advancement in retrieval-augmented generation techniques.
The paper tackles the problem of managing user data from personal devices to create personalized agents, achieving about a 10% improvement over existing methods in experiments and enhancing usability in a real smartphone assistant.
In the age of mobile internet, user data, often referred to as memories, is continuously generated on personal devices. Effectively managing and utilizing this data to deliver services to users is a compelling research topic. In this paper, we introduce a novel task of crafting personalized agents powered by large language models (LLMs), which utilize a user's smartphone memories to enhance downstream applications with advanced LLM capabilities. To achieve this goal, we introduce EMG-RAG, a solution that combines Retrieval-Augmented Generation (RAG) techniques with an Editable Memory Graph (EMG). This approach is further optimized using Reinforcement Learning to address three distinct challenges: data collection, editability, and selectability. Extensive experiments on a real-world dataset validate the effectiveness of EMG-RAG, achieving an improvement of approximately 10% over the best existing approach. Additionally, the personalized agents have been transferred into a real smartphone AI assistant, which leads to enhanced usability.