Evaluating Memory Capability in Continuous Lifelog Scenario

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zheng, Jianjie, Liu, Zhichen, Shen, Zhanyu, Qu, Jingxiang, Chen, Guanhua, Wang, Yile, Xu, Yang, Liu, Yang, Cheng, Sijie
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914481780555776
author Zheng, Jianjie
Liu, Zhichen
Shen, Zhanyu
Qu, Jingxiang
Chen, Guanhua
Wang, Yile
Xu, Yang
Liu, Yang
Cheng, Sijie
author_facet Zheng, Jianjie
Liu, Zhichen
Shen, Zhanyu
Qu, Jingxiang
Chen, Guanhua
Wang, Yile
Xu, Yang
Liu, Yang
Cheng, Sijie
contents Nowadays, wearable devices can continuously lifelog ambient conversations, creating substantial opportunities for memory systems. However, existing benchmarks primarily focus on online one-on-one chatting or human-AI interactions, thus neglecting the unique demands of real-world scenarios. Given the scarcity of public lifelogging audio datasets, we propose a hierarchical synthesis framework to curate \textbf{\textsc{LifeDialBench}}, a novel benchmark comprising two complementary subsets: \textbf{EgoMem}, built on real-world egocentric videos, and \textbf{LifeMem}, constructed using simulated virtual community. Crucially, to address the issue of temporal leakage in traditional offline settings, we propose an \textbf{Online Evaluation} protocol that strictly adheres to temporal causality, ensuring systems are evaluated in a realistic streaming fashion. Our experimental results reveal a counterintuitive finding: current sophisticated memory systems fail to outperform a simple RAG-based baseline. This highlights the detrimental impact of over-designed structures and lossy compression in current approaches, emphasizing the necessity of high-fidelity context preservation for lifelog scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2604_11182
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Evaluating Memory Capability in Continuous Lifelog Scenario
Zheng, Jianjie
Liu, Zhichen
Shen, Zhanyu
Qu, Jingxiang
Chen, Guanhua
Wang, Yile
Xu, Yang
Liu, Yang
Cheng, Sijie
Computation and Language
Nowadays, wearable devices can continuously lifelog ambient conversations, creating substantial opportunities for memory systems. However, existing benchmarks primarily focus on online one-on-one chatting or human-AI interactions, thus neglecting the unique demands of real-world scenarios. Given the scarcity of public lifelogging audio datasets, we propose a hierarchical synthesis framework to curate \textbf{\textsc{LifeDialBench}}, a novel benchmark comprising two complementary subsets: \textbf{EgoMem}, built on real-world egocentric videos, and \textbf{LifeMem}, constructed using simulated virtual community. Crucially, to address the issue of temporal leakage in traditional offline settings, we propose an \textbf{Online Evaluation} protocol that strictly adheres to temporal causality, ensuring systems are evaluated in a realistic streaming fashion. Our experimental results reveal a counterintuitive finding: current sophisticated memory systems fail to outperform a simple RAG-based baseline. This highlights the detrimental impact of over-designed structures and lossy compression in current approaches, emphasizing the necessity of high-fidelity context preservation for lifelog scenarios.
title Evaluating Memory Capability in Continuous Lifelog Scenario
topic Computation and Language
url https://arxiv.org/abs/2604.11182