2024/08/18 by Kaicheng Yang, Yang, Kaicheng, T. Gu +13 · 9 citations
Biochemistry, Genetics and Molecular Biology · Chemistry · Engineering · #Advanced Control Systems Optimization #Chemical Synthesis and Reactions #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Phytochemical Studies and Bioactivities
paper · pdf · doi:10.48550/arxiv.2408.09441
openalex publication_date 2024/08/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Contrastive Language-Image Pre-training (CLIP) has achieved excellent performance over a wide range of tasks. However, the effectiveness of CLIP heavily relies on a substantial corpus of pre-training data, resulting in notable consumption of computational resources. Although knowledge distillation has been widely applied in single modality models, how to efficiently expand knowledge distillation to vision-language foundation models with extensive data remains relatively unexplored. In this paper, we introduce CLIP-CID, a novel distillation mechanism that effectively transfers knowledge from a large vision-language foundation model to a smaller model. We initially propose a simple but efficient image semantic balance method to reduce transfer learning bias and improve distillation efficiency. This method filters out 43.7% of image-text pairs from the LAION400M while maintaining superior performance. After that, we leverage cluster-instance discrimination to facilitate knowledge transfer from the teacher model to the student model, thereby empowering the student model to acquire a holistic semantic comprehension of the pre-training data. Experimental results demonstrate that CLIP-CID achieves state-of-the-art performance on various downstream tasks including linear probe and zero-shot classification.