GP-Adapter enhances CLIP for few-shot classification and OOD detection

Researchers have introduced GP-Adapter, a novel framework designed to enhance CLIP's capabilities for few-shot classification and out-of-distribution detection. This method integrates Gaussian Process uncertainty modeling with CLIP's pre-trained embeddings without requiring any fine-tuning of the base model. By constructing class-wise GPs on frozen CLIP features, GP-Adapter generates variance-aware confidence scores, proving effective in low-data and distribution-shifted scenarios. AI

IMPACT Enhances reliability of vision-language models in low-data and distribution-shifted settings.

RANK_REASON This is a research paper describing a new method for improving existing models. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Gaussian Process
• GP-Adapter
• ImageNet

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →