Post

RTVLM

  • Paper: M3IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning
  • GitHub Link
  • Publisher: Arxiv
  • Author Affiliation: The University of Hong Kong
  • Type
    • SFT
    • RLHF
  • Multi-turn
  • Input Modalities $\rightarrow$ Output Modalities
    (I: Image, V: Video, A: Audio, 3D: Point Cloud, T: Text, B: Bounding box, Tab: Table, Web: Web page)
    • I+T $\rightarrow$ T
  • Source
    • New question-image pairs based on publicly available images or originally diffusion-generated images.
  • Method
    • Auto. + Manu.
  • I/V/A Scale
    • I
      • Not report
    • V
      • Not report
    • A
      • Not report
  • Dialog Turn
    • Not report
  • Instance Scale
    • 5K
This post is licensed under CC BY 4.0 by the author.