ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision

  • 类型:arxiv
  • 标识:2102.03334
  • 链接:https://arxiv.org/abs/2102.03334
  • 主题:multimodal
  • 主分类:multimodal
  • 形态:method
  • 被引:2389
  • 被引来源:Semantic Scholar
  • S2被引:2389
  • OpenAlex被引:539
  • 影响力被引:230
  • TLDR:A minimal VLP model, Vision-and-Language Transformer (ViLT), monolithic in the sense that the processing of visual inputs is drastically simplified to just the same convolution-free manner that the authors process textual inputs, showing that ViLT is up to tens of times faster than previous VLP models, yet with competitive or better downstream task performance.
  • OpenAlex ID:W3126792443
  • OpenAlex DOI:10.48550/arxiv.2102.03334
  • DOI:10.48550/arxiv.2102.03334
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://arxiv.org/pdf/2102.03334
  • OpenAlex更新:2026-08-23
  • 待LLM分类:否
  • 标题中文:ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision
  • TLDR中文:提出极简的 VLP 模型 Vision-and-Language Transformer (ViLT),其一体化设计将视觉输入处理大幅简化为与文本输入相同的无卷积方式;ViLT 比此前的 VLP 模型快达数十倍,同时下游任务性能具有竞争力甚至更优。
  • 来源文件
  • [OpenAlex discover]
  • [S2 enrich]