Mem-Gallery
Mem-Gallery: Benchmarking Multimodal Long-Term Conversational Memory for MLLM Agents
Yuanchen Bei, Tianxin Wei, Xuying Ning, Yanjun Zhao, Zhining Liu, Xiao Lin, et al. · 2026-01-07
- Modality
- Image
- Data
- Synthetic
- Task · ground truth
- QA/Ret. · Ans.+Ev.
- Scope
- Memory-specific benchmark
What it tests
- CECross-modal evidenceDirectly evaluated
- LELong-range evidenceDirectly evaluated
- SCState changesDirectly evaluated
- CFConflicting memoriesDirectly evaluated
- UQUnanswerable queriesDirectly evaluated
- MAMemory-grounded actionNot evaluated
- FGFine-grained visual detailPartially evaluated
- ETEntity trackingPartially evaluated
- VCVisual state changeDirectly evaluated
Survey notes
Memory accumulates across multimodal dialogue sessions and is tested for factual and visual recall, temporal reasoning, updates and inconsistent or insufficient evidence.