Skip to content
Hot storiesContinuously updated

华中科大等提出全连续多模态架构Multimodal Flow

1 reports1 reporting sourcesUpdated 3 hours ago

Start here

AI overview

2026年11月6日，据报道，华中科技大学与地平线机器人等机构联合提出全连续多模态架构 Multimodal Flow，以实现统一语言与视觉流建模。该架构将语言和视觉分别保留在各自的连续嵌入空间中，并共享统一的 Flow Matching 目标来建模二者的联合分布。

AI-generated from reports · Updated 3 hours ago

Latest developmentsNov 6 10:51
华中科大与地平线机器人等提出Multimodal Flow，统一语言与视觉流建模。

Report timeline

Follow the reports to see different sides of the story.

Nov 6
  1. AITNT · AI头条
    华中科大等提出全连续多模态架构 Multimodal Flow，统一语言与视觉流建模

    华中科技大学与地平线机器人等机构提出全连续多模态架构 Multimodal Flow，将语言和视觉保留在各自的连续嵌入空间中，共享统一的 Flow Matching 目标建模联合分布。

Story heat trend

Current heat 9·Comparable peak 10（Nov 6 11:00）·Comparable change, last 24 hours–

02.557.510Nov6Nov6Nov6Nov6

Trends compare only the same parties observed continuously, so the scope can be narrower than current heat counts. Move the pointer or tap the chart to see hourly heat; use the left and right arrow keys on a keyboard.