Scholar Feed

Agent / long-term memory

Mem-Gallery

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

Superseded baseline#56 of 63 most-superseded · first seen Jan 7, 2026

Cited as a baseline — critiqued by newer work, not yet beaten on a benchmark here

1 papers critique it · 0 beat it on benchmarks

What papers say

Verbatim critique sentences, each from a paper that cites Mem-Gallery as a baseline.

Multimodal conversational benchmarks such as LoCoMo~maharana2024evaluatinglongtermconversationalmemory and Mem-Gallery~memgallery retain both visual and text modalities, yet most of their questions admit a text-only shortcut, rendering the visual modality redundant.
— MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.