visual

Definition

An adjective for image- and video-derived information, especially in tasks that tie perception to language: visual question answering, visual grounding, visual instruction tuning. In practice "vision" tends to name the pure perception field, while "visual" more often appears where the visual channel is one input among several. The distinction is soft and usage varies between subcommunities.

Explainers using this term

• Paper Explained: SenseNova-U1.5 — One Model That Both Sees and Draws, With No Encoder and No VAESenseNova-U1.5: Towards Native Unified Visual Intelligence
• Paper Explained: Reason Through the Latent! — Visual information being in a latent state doesn't mean the answer uses itReason Through the Latent! Making Latent Visual Reasoning Necessary
• Editable Visual Design, explained — demoting image generation to a sketch and shipping the artifact as HTMLEditable Visual Design