artificial intelligence

LightMem-Ego: Your AI Memory for Everyday Life

arXiv:2607.11487

summary

The paper introduces LightMem-Ego, a lightweight on‑device system that continuously records egocentric video and audio, organizes them into hierarchical short‑ and long‑term memories, and retrieves relevant multimodal evidence to answer user queries about past experiences.

Abstract

Personal AI assistants on mobile and wearable devices continuously perceive users' daily lives through visual and audio streams. However, answering queries about past experiences requires lightweight multimodal memory that can continuously accumulate, organize, and retrieve long-term experiences, which remains challenging. To address this challenge, we present LightMem-Ego, a lightweight streaming multimodal memory system for everyday-life assistance. The system continuously captures egocentric visual and audio streams, aligns them on a shared timeline, and organizes them into a hierarchical memory consisting of current, short-term, and long-term memory. Given a user query, LightMem-Ego dynamically routes retrieval to the appropriate memory level and generates answers grounded in multimodal evidence. The demonstration can be deployed on smartphones and AI glasses, supporting object finding, conversation recall, life summarization, routine discovery, and personalized assistance. Code is available at https://github.com/zjunlp/LightMem-Ego.

Ongoing work

Topics & keywords

#egocentric memory#multimodal streaming#personal AI assistants#hierarchical memory#on‑device inferenceegocentric videoaudio‑visual alignmentlightweight multimodal memoryhierarchical retrievalmobile AI deployment