AI Paper Atlas

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

Probabilistic Prompt Learning for Dense Prediction

Hyeongjun Kwon, Taeyong Song, Somi Jeong, Jin Kim, Jinhyun Jang, Kwanghoon Sohn

CVPR 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Prompt learning for CLIP · 22 citations (OpenAlex)

Read the paper

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
Use probabilistic, visual-context-aware prompts instead of deterministic promptsSemantic FPN · ADE20k valmIoU-SS: 43.5 -> 44.7Improved
Use probabilistic, visual-context-aware prompts instead of deterministic promptsMask R-CNN, ResNet-50 · COCO valAP^b: 40.2 -> 41.0Improved
Use probabilistic, visual-context-aware prompts instead of deterministic promptsMask R-CNN, ResNet-50 · COCO valAP^m: 37.6 -> 38.3Improved
Model prompt distributions with visual-text context rather than text-only prompt statisticsSemantic FPN, ResNet-50 · ADE20k valmIoU-SS: 42.6 -> 44.7Improved
Add diversity loss (ablation)PPL · ADE20k valmIoU: 42.0 -> 42.4Improved
Add KL divergence loss (ablation)PPL · ADE20k valmIoU: 42.0 -> 43.6Improved
Add diversity loss and KL divergence loss together (ablation)PPL · ADE20k valmIoU: 42.0 -> 44.3Improved
Replace pixel-matching loss with probabilistic pixel-text matching loss (ablation)PPL with diversity and KL losses · ADE20k valmIoU: 44.3 -> 44.7Improved
Use 3 attribute prompts instead of 1 (ablation)PPL · ADE20k valmIoU: 43.8 -> 44.7Improved
Increase attribute prompts from 3 to 5 (ablation)PPL · ADE20k valmIoU: 44.7 -> 44.2Hurt
Set KL-divergence weight β to 10⁻⁷ instead of 10⁻⁵ (ablation)PPL · ADE20k valmIoU: 44.7 -> 42.4Hurt
Set KL-divergence weight β to 10⁻⁴ instead of 10⁻⁵ (ablation)PPL · ADE20k valmIoU: 44.7 -> 44.1Hurt

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