MetaCLIP is a research codebase that extends the CLIP framework into a meta-learning / continual learning regime, aiming to adapt CLIP-style models to new tasks or domains efficiently. The goal is to preserve CLIP’s strong zero-shot transfer capability while enabling fast adaptation to domain shifts or novel class sets with minimal data and without catastrophic forgetting. The repository provides training logic, adaptation strategies (e.g. prompt tuning, adapter modules), and evaluation across base and target domains to measure how well the model retains its general knowledge while specializing as needed. It includes utilities to fine-tune vision-language embeddings, compute prompt or adapter updates, and benchmark across transfer and retention metrics. MetaCLIP is especially suited for real-world settings where a model must continuously incorporate new visual categories or domains over time.
Features
- Meta-learning extensions to CLIP for efficient domain adaptation
- Prompt tuning or adapter modules for lightweight specialization
- Evaluation protocols to test transfer and retention (avoiding forgetting)
- Tools for fine-tuning vision-text embeddings on new data
- Modular architecture separating base CLIP and adaptation layers
- Scripts and benchmarks for continual adaptation across domains
Project Samples
