Mem-Gallery: Benchmarking Multimodal Long-Term Conversational Memory for MLLM Agents

TL;DR

Mem-Gallery introduces a comprehensive benchmark for evaluating multimodal long-term conversational memory in MLLM agents, addressing gaps in existing assessments by focusing on long-term, multi-session multimodal interactions.

Contribution

The paper presents Mem-Gallery, a new benchmark dataset and evaluation framework for assessing long-term multimodal memory in MLLM agents, highlighting key challenges and limitations.

Findings

Explicit multimodal memory retention is essential.

Current models struggle with memory reasoning.

Memory organization impacts long-term performance.

Abstract

Long-term memory is a critical capability for multimodal large language model (MLLM) agents, particularly in conversational settings where information accumulates and evolves over time. However, existing benchmarks either evaluate multi-session memory in text-only conversations or assess multimodal understanding within localized contexts, failing to evaluate how multimodal memory is preserved, organized, and evolved across long-term conversational trajectories. Thus, we introduce Mem-Gallery, a new benchmark for evaluating multimodal long-term conversational memory in MLLM agents. Mem-Gallery features high-quality multi-session conversations grounded in both visual and textual information, with long interaction horizons and rich multimodal dependencies. Building on this dataset, we propose a systematic evaluation framework that assesses key memory capabilities along three functional dimensions: memory extraction and test-time adaptation, memory reasoning, and memory knowledge management. Extensive benchmarking across thirteen memory systems reveals several key findings, highlighting the necessity of explicit multimodal information retention and memory organization, the persistent limitations in memory reasoning and knowledge management, as well as the efficiency bottleneck of current models.