{"id":"581611f3-45fb-4eb6-8ad4-13c20898b2d6","kind":"products","code":"UB-D931","name":"Mingjingduomotaidamoxing","nameBasis":"transliterated","summary":"We propose a new multimodal large model paradigm: the Hypergraph Multimodal Large Language Model (HMLLM). It uses hypergraphs to model complex relationships among video elements, EEG signals, and eye-tracking data. Integrating modalities closes semantic gaps and improves reasoning and semantic analysis, allowing HMLLM to simulate different groups' varied responses to the same video elements.","description":"We propose a new multimodal large model paradigm: the Hypergraph Multimodal Large Language Model (HMLLM). It uses hypergraphs to model complex relationships among video elements, EEG signals, and eye-tracking data. Integrating modalities closes semantic gaps and improves reasoning and semantic analysis, allowing HMLLM to simulate different groups' varied responses to the same video elements.","country":null,"category":"Video generation","originalName":"明敬多模态大模型","eventDate":null,"website":null,"sourceUpdatedAt":"2026-10-18T17:06:10.412Z","englishUpdatedAt":"2026-10-20T18:30:06.661Z","reviewedAt":"2026-10-20T18:30:06.661Z","snapshotAt":"2026-10-20T18:06:34.551Z","sourceHash":"412e76e88057082b7562bf4a76fd28efb9fdd6382ef8957e8105d2022229fe04","companyId":"da434cea-215f-442a-a242-43da22894262","foundedYear":null,"logoUrl":"https://c.ux-cdn.com/1863156737900573-5aba7b40bf5e48ca8be5896e57b70eca/uniquebase%2Flogos%2Fproduct%2FrecTXQ8pCP%2F7afa355ce20f0c367d69d559925bf715af41512f909018e41309c1a504d80c48.png","originalText":"我们提出了全新的多模态大模型的范式：超图多模态大语言模型（HMLLM），它利用超图构建视频元素、脑电信号和眼动追踪数据之间的复杂关系。通过整合不同模态的信息，HMLLM弥合了丰富模态间的语义差距，增强了其逻辑推理和语义分析的能力。使得HMLLM可以模拟不同人群观看到同一段视频元素时的异质反应。","sources":[],"related":[{"id":"da434cea-215f-442a-a242-43da22894262","kind":"companies","code":"UB-C1402","name":"MiningLamp","nameBasis":"source-english","summary":"Enterprise data analytics and organizational intelligence services platform.","description":"Mininglamp Technology (2718.HK), founded in 2006, is a leading Chinese agentic service company with internally developed model capabilities. In 2025, it listed on the Hong Kong Stock Exchange as the world's first agentic AI stock.","country":"China","category":null,"originalName":"明略科技","eventDate":null,"website":"https://www.mininglamp.com/","sourceUpdatedAt":"2026-10-05T00:00:00.000Z","englishUpdatedAt":"2026-10-20T18:26:36.712Z","reviewedAt":"2026-10-20T18:26:36.712Z","snapshotAt":"2026-10-20T18:05:01.520Z","sourceHash":"6134b9ac7fac97636678b08b11f2eeb7218fbac66dacd8fae88dd0c2db108555","companyId":null,"foundedYear":2006,"logoUrl":"https://c.ux-cdn.com/1863156737900573-5aba7b40bf5e48ca8be5896e57b70eca/uniquebase%2Flogos%2Fcompany%2FreciHQY9aM%2Fcf9d0754a65789f8f6303e8bd0a2c32f958d204e7ac9d00f1bb2dacc863a8a2f.png","path":"/companies/minglvekeji"}],"path":"/products/mingjingduomotaidamoxing","structured":{"city":null,"country":null,"location":null,"role":null,"category":"Video generation","company":{"name":"MiningLamp","path":"/companies/minglvekeji"}},"definition":"Mingjingduomotaidamoxing is a product in the Video generation category from MiningLamp.","summaryBasis":"translated","thin":false,"canonical":"https://uniquebase.ai/products/mingjingduomotaidamoxing"}