LightMem-Ego: Your AI Memory for Everyday Life

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Yijun Chen, Boyi Xiao, Yixian Zhao, Haoting Xia, Buqiang Xu, Jizhan Fang, Yanya Li, Yaqi Zheng, Xuehai Wang, Zirui Xue, Liuxin Zhang, Hui Li, Ningyu Zhang
• Venue: arXiv: cs.CL
• Published: 2026-07-13
• Collected: 2026-07-14
• URL: http://arxiv.org/abs/2607.11487v1
• PDF: https://arxiv.org/pdf/2607.11487v1
• DOI: UNKNOWN

Plain-Language Brief

The research paper introduces LightMem-Ego, a lightweight multimodal memory system designed for AI assistants on mobile and wearable devices. The system continuously captures and organizes egocentric visual and audio streams, creating a hierarchical memory structure (current, short-term, long-term). It can retrieve information based on user queries and provide answers grounded in multimodal evidence. The system is designed to support various tasks such as object finding, conversation recall, life summarization, routine discovery, and personalized assistance.

What the Document Says

• The research paper presents LightMem-Ego, a lightweight streaming multimodal memory system for everyday-life assistance.
• The system captures egocentric visual and audio streams continuously.
• It organizes the captured data 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 system supports tasks like object finding, conversation recall, life summarization, routine discovery, and personalized assistance.
• The code for LightMem-Ego is available on GitHub.

Key Concepts Mentioned

• LightMem-Ego
• AI assistant
• egocentric visual and audio streams
• hierarchical memory
• user query
• multimodal evidence
• object finding
• conversation recall
• life summarization
• routine discovery
• personalized assistance

Claims Made by the Document

• LightMem-Ego is a lightweight streaming multimodal memory system.
• The system can continuously capture egocentric visual and audio streams.
• It organizes the captured data into a hierarchical memory structure.
• Given a user query, LightMem-Ego generates answers grounded in multimodal evidence.
• The system supports various tasks like object finding, conversation recall, life summarization, routine discovery, and personalized assistance.

Evidence Used

• The text states that the research paper introduces LightMem-Ego as a lightweight streaming multimodal memory system for everyday-life assistance.
• It mentions that the system captures egocentric visual and audio streams continuously.
• The system is described as organizing the captured data into a hierarchical memory structure.
• The text states that given a user query, LightMem-Ego generates answers grounded in multimodal evidence.
• The paper claims that the system supports various tasks like object finding, conversation recall, life summarization, routine discovery, and personalized assistance.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

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.