Guardado en:
Detalles Bibliográficos
Autores principales: Feng, Junyu, Xu, Binxiao, Chen, Jiayi, Dai, Mengyu, Wu, Cenyang, Li, Haodong, Zeng, Bohan, Xie, Yunliu, Liang, Hao, Lu, Ming, Zhang, Wentao
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2602.07624
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910015357452288
author Feng, Junyu
Xu, Binxiao
Chen, Jiayi
Dai, Mengyu
Wu, Cenyang
Li, Haodong
Zeng, Bohan
Xie, Yunliu
Liang, Hao
Lu, Ming
Zhang, Wentao
author_facet Feng, Junyu
Xu, Binxiao
Chen, Jiayi
Dai, Mengyu
Wu, Cenyang
Li, Haodong
Zeng, Bohan
Xie, Yunliu
Liang, Hao
Lu, Ming
Zhang, Wentao
contents This work addresses the challenge of personalized question answering in long-term human-machine interactions: when conversational history spans weeks or months and exceeds the context window, existing personalization mechanisms struggle to continuously absorb and leverage users' incremental concepts, aliases, and preferences. Current personalized multimodal models are predominantly static-concepts are fixed at initialization and cannot evolve during interactions. We propose M2A, an agentic dual-layer hybrid memory system that maintains personalized multimodal information through online updates. The system employs two collaborative agents: ChatAgent manages user interactions and autonomously decides when to query or update memory, while MemoryManager breaks down memory requests from ChatAgent into detailed operations on the dual-layer memory bank, which couples a RawMessageStore (immutable conversation log) with a SemanticMemoryStore (high-level observations), providing memories at different granularities. In addition, we develop a reusable data synthesis pipeline that injects concept-grounded sessions from Yo'LLaVA and MC-LLaVA into LoCoMo long conversations while preserving temporal coherence. Experiments show that M2A significantly outperforms baselines, demonstrating that transforming personalization from one-shot configuration to a co-evolving memory mechanism provides a viable path for high-quality individualized responses in long-term multimodal interactions. The code is available at https://github.com/Little-Fridge/M2A.
format Preprint
id arxiv_https___arxiv_org_abs_2602_07624
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle M2A: Multimodal Memory Agent with Dual-Layer Hybrid Memory for Long-Term Personalized Interactions
Feng, Junyu
Xu, Binxiao
Chen, Jiayi
Dai, Mengyu
Wu, Cenyang
Li, Haodong
Zeng, Bohan
Xie, Yunliu
Liang, Hao
Lu, Ming
Zhang, Wentao
Artificial Intelligence
This work addresses the challenge of personalized question answering in long-term human-machine interactions: when conversational history spans weeks or months and exceeds the context window, existing personalization mechanisms struggle to continuously absorb and leverage users' incremental concepts, aliases, and preferences. Current personalized multimodal models are predominantly static-concepts are fixed at initialization and cannot evolve during interactions. We propose M2A, an agentic dual-layer hybrid memory system that maintains personalized multimodal information through online updates. The system employs two collaborative agents: ChatAgent manages user interactions and autonomously decides when to query or update memory, while MemoryManager breaks down memory requests from ChatAgent into detailed operations on the dual-layer memory bank, which couples a RawMessageStore (immutable conversation log) with a SemanticMemoryStore (high-level observations), providing memories at different granularities. In addition, we develop a reusable data synthesis pipeline that injects concept-grounded sessions from Yo'LLaVA and MC-LLaVA into LoCoMo long conversations while preserving temporal coherence. Experiments show that M2A significantly outperforms baselines, demonstrating that transforming personalization from one-shot configuration to a co-evolving memory mechanism provides a viable path for high-quality individualized responses in long-term multimodal interactions. The code is available at https://github.com/Little-Fridge/M2A.
title M2A: Multimodal Memory Agent with Dual-Layer Hybrid Memory for Long-Term Personalized Interactions
topic Artificial Intelligence
url https://arxiv.org/abs/2602.07624