Research map / Vision-language and multimodal / CLIP and zero-shot
UP-DP: Unsupervised Prompt Learning for Data Pre-Selection with Vision-Language Models
Xin Li, Sima Behpour, Thang Doan, Wenbin He, Liang Gou, Liu Ren
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use UP-DP cluster-head selection instead of USL-M k-means selection on same learned-prompt features | UP-DP · 7-dataset average, CLIP ViTG14 | accuracy: 0.300 -> 0.371 | Improved |
| use multimodal features with learned prompt (USL-M) instead of image features (USL-I) | USL · 7-dataset average, CLIP ViTG14 | accuracy: 0.290 -> 0.300 | Improved |
| shorten learnable context from 16 to 4 tokens (ablation) | UP-DP · 7-dataset average, CLIP ViTG14 | accuracy: 0.322 -> 0.354 | Improved |
| shorten learnable context from 16 to 4 tokens (ablation) | UP-DP · 7-dataset average, CLIP RN50 | accuracy: 0.210 -> 0.240 | Improved |
| apply prompt learned on Flowers102 to all datasets instead of image-only features | BLIP-2 KNN features · 7-dataset average KNN | accuracy: 0.760 -> 0.864 | Improved |
| apply prompt learned on EuroSAT instead of image-only features | BLIP-2 KNN features · OxfordPets KNN | accuracy: 0.681 -> 0.501 | Hurt |
| use empty prompt instead of image-only features | BLIP-2 KNN features · 7-dataset average KNN | accuracy: 0.760 -> 0.606 | Hurt |
| use initial (untrained) prompt instead of image-only features | BLIP-2 KNN features · 7-dataset average KNN | accuracy: 0.760 -> 0.592 | Hurt |
| use cluster head g_C features instead of backbone f features (ablation) | UP-DP · 7-dataset average, CLIP ViTG14 | accuracy: 0.354 -> 0.371 | Improved |
| use instance head g_I features instead of backbone f features (ablation) | UP-DP · 7-dataset average, CLIP ViTG14 | accuracy: 0.354 -> 0.357 | No clear effect |
| increase annotation budget by 50 percent (300 labels) | UP-DP vs Random · 7-dataset average, CLIP ViTG14 | accuracy: 0.400 -> 0.423 | Mixed |
| UP-DP versus random selection on Caltech101 | UP-DP · Caltech101, CLIP ViTG14 | accuracy: 0.334 -> 0.339 | No clear effect |
Closest papers by meaning
- Make Prompts Adaptable: Bayesian Modeling for Vision-Language Prompt Learning with Data-Dependent Prior (AAAI 2024 · similarity 0.88)
- ProAPO: Progressively Automatic Prompt Optimization for Visual Classification (CVPR 2025 · similarity 0.87)
- Tree of Attributes Prompt Learning for Vision-Language Models (ICLR 2025 · similarity 0.87)
- Beyond Words: Augmenting Discriminative Richness via Diffusions in Unsupervised Prompt Learning (CVPR 2025 · similarity 0.87)
- Weighted Multi-Prompt Learning with Description-free Large Language Model Distillation (ICLR 2025 · similarity 0.87)
- NLPrompt: Noise-Label Prompt Learning for Vision-Language Models (CVPR 2025 · similarity 0.87)
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.
