Evaluating Memory Capability in Continuous Lifelog Scenario

TL;DR

This paper introduces extsc{LifeDialBench}, a new benchmark for evaluating memory systems in continuous lifelog scenarios, emphasizing realistic online evaluation and revealing limitations of current methods.

Contribution

It presents a hierarchical synthesis framework for creating lifelogging benchmarks and proposes an online evaluation protocol that prevents temporal leakage.

Findings

Current memory systems do not outperform simple RAG baselines.

Over-designed structures and lossy compression harm performance.

High-fidelity context preservation is crucial for lifelog scenarios.

Abstract

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.