Visual Instruction Tuning Aligns Modalities through Abstraction 文章

ArXiv CS.CV2026-06-03NEWSen作者: Luis Palacios, Lorenzo Basile, Diego Doimo, Alberto Cazzaniga
查看原文 →
关系图谱

Visual Instruction Tuning Aligns Modalities through Abstraction · 相关技术

相关技术

vision-language architecturesLarge language model (LLM)instruction tuning