arXiv cs.CL· Jielong Tang, Ling-An Zeng, Xujie Yuan, Xiao Dong, Zhiqian Xia, Qingfeng Li, Shixiong Kai, Kaichao Liang, Rong Yao, Jian Yin·· 13 小时前AI 评分31
MHG-Mem：为多模态智能体保留溯源信息的超图记忆
Remembering Through Change: Provenance-Preserving Hypergraph Memory for Multimodal Agents
AI 导读
研究者提出多模态超图记忆 MHG-Mem，将实体提及与观测事实表示为节点，通过出现超边聚合身份一致的提及，并用演化超边和证据超边关联状态变化与视觉源证据。配套的超图智能体搜索可自适应选择入口层、跨层遍历类型化超边并按需核查原始观测。在三个基准上，MHG-Mem 相比最强基线提升 8.20-9.98%。
正文 · 原文
Abstract:Long-term memory for multimodal agents remains challenging, as it must preserve fine-grained visual evidence while tracking entity evolution across sessions. Existing abstraction-based memories compress images into text, often discarding visually discriminative details, whereas pixel-preserving memories retain raw observations but struggle to reliably associate diverse observations with persistent entity identities and track their evolving states. To address these limitations, we propose MHG-Mem, a multimodal hypergraph memory that jointly organizes entity identity, state evolution, and provenance to visual evidence. Specifically, MHG-Mem represents distinct entity mentions and their observed facts as individual nodes, grouping identity-consistent mentions and facts through occurrence hyperedges. It then synthesizes aspect-specific evolution states from these grouped observations and links them to supporting entities, facts, and source evidence through evolutionary and evidential hyperedges. To navigate this structured memory, we introduce hypergraph-based agentic search, which adaptively selects an entry layer, traverses typed hyperedges across memory levels, and inspects source observations when fine-grained verification is required. These designs support both detailed visual verification and reasoning over cross-session state changes. Extensive experiments across three benchmarks demonstrate the effectiveness of MHG-Mem, improving performance by 8.20-9.98% over the strongest baseline.
| Comments: | 31 pages,10 figures |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.29995 [cs.CL] |
| (or arXiv:2610.29995v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.29995 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jielong Tang [view email]
[v1]
Wed, 21 Oct 2026 16:34:28 UTC (4,080 KB)
来源：arXiv cs.CL · arxiv.org