the.ai

Multimodal / Objectives

verified

CLIP

CLIP trains an image encoder and a text encoder together so that a picture and its caption land in the same place. Once they share a space, classification needs no classifier: encode the candidate labels as sentences and pick whichever sits closest. It learns from captions scraped at scale rather than from anyone labelling images by hand.

Viz primitive · attention-heatmaptokens = 8
query ↓key →

8 queries against 8 keys; a brighter cell means more of that query's attention went to that key. Each query sees itself and everything before it, and nothing after — the upper triangle is masked.

Image-text similarity across a batch, with matched pairs on the diagonal. Drag the batch size to watch the negatives multiply.

8

Reviewed by opendroid · 2026-08-04

  • arXiv:2103.00020 — Learning Transferable Visual Models From Natural Language Supervision