MMA通过动态评分提升多模态记忆可靠性，减少错误推理。
MMA: Multimodal Memory Agent
- 用可信度、时间衰减和冲突共识动态打分记忆项
- 在FEVER上准确率持平但方差降35.2%，在MMA-Bench上视觉模式达41.18%准确率
- 适合关注多模态推理鲁棒性与安全性的研究者
长时程多模态智能体依赖外部记忆；然而基于相似性的检索常召回过时、低可信或冲突的条目，引发过度自信的错误。我们提出多模态记忆代理（MMA），通过融合源可信度、时间衰减和冲突感知网络共识，为每个检索记忆项分配动态可靠性分数，并据此重加权证据或在支持不足时放弃判断。我们还引入MMA-Bench，一个程序生成的基准测试，用于评估信念演化，具备可控的说话人可信度和结构化的文本-视觉矛盾。在此框架下，我们揭示了“视觉安慰剂效应”，发现RAG类智能体会继承基础模型中的潜在视觉偏见。在FEVER上，MMA保持基线准确率的同时将方差降低35.2%，并提升选择性效用；在LoCoMo上，安全配置提升了可操作准确率并减少了错误回答；在MMA-Bench上，MMA在视觉模式下达到41.18%的Type-B准确率，而基线则降至0.0%。代码已开源。
原文摘要 · Abstract (English)
Long-horizon multimodal agents depend on external memory; however, similarity-based retrieval often surfaces stale, low-credibility, or conflicting items, which can trigger overconfident errors. We propose Multimodal Memory Agent (MMA), which assigns each retrieved memory item a dynamic reliability score by combining source credibility, temporal decay, and conflict-aware network consensus, and uses this signal to reweight evidence and abstain when support is insufficient. We also introduce MMA-Bench, a programmatically generated benchmark for belief dynamics with controlled speaker reliability and structured text-vision contradictions. Using this framework, we uncover the "Visual Placebo Effect", revealing how RAG-based agents inherit latent visual biases from foundation models. On FEVER, MMA matches baseline accuracy while reducing variance by 35.2% and improving selective utility; on LoCoMo, a safety-oriented configuration improves actionable accuracy and reduces wrong answers; on MMA-Bench, MMA reaches 41.18% Type-B accuracy in Vision mode, while the baseline collapses to 0.0% under the same protocol. Code: https://github.com/AIGeeksGroup/MMA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。