AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

LaFTer: Label-Free Tuning of Zero-shot Classifier using Language and Unlabeled Image Collections

Muhammad Jehanzeb Mirza, Leonid Karlinsky, Wei Lin, Horst Possegger, Mateusz Kozinski, Rogerio Feris, Horst Bischof

NeurIPS 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Adapting VLMs for zero-shot classification · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply LaFTer label-free tuningCLIP ViT-B/32 · EuroSattop-1 accuracy: 45.1 -> 73.9Improved
apply LaFTer label-free tuningCLIP ViT-B/32 · ImageNettop-1 accuracy: 61.9 -> 64.2Improved
use LaFTer instead of UPLUPL · EuroSattop-1 accuracy: 62.2 -> 73.9Improved
use LaFTer instead of UPLUPL · DTDtop-1 accuracy: 48.0 -> 46.1Hurt
use LaFTer instead of UPLUPL · ImageNet-Atop-1 accuracy: 26.9 -> 31.5Improved
remove learnable visual prompts (ablation)LaFTer · 7-dataset averagetop-1 accuracy: 61.4 -> 65.6Improved
add the classifier (instead of CLIP cosine similarity in both branches) (ablation)LaFTer · 7-dataset averagetop-1 accuracy: 60.4 -> 65.6Improved
add augmentations (ablation)LaFTer · 7-dataset averagetop-1 accuracy: 61.8 -> 65.6Improved
add affine parameter tuning (ablation)LaFTer · 7-dataset averagetop-1 accuracy: 62.0 -> 65.6Improved
add stop gradient (ablation)LaFTer · 7-dataset averagetop-1 accuracy: 64.2 -> 65.6Improved
add text-only pre-training (ablation)LaFTer · 7-dataset averagetop-1 accuracy: 63.7 -> 65.6Improved
add text-only pre-training (ablation)LaFTer · EuroSattop-1 accuracy: 69.7 -> 73.9Improved

Later papers that compared with this method

2 later accepted papers report LafTer as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy