DiffCLIP: Differential Attention Meets CLIP

TMLR Paper4934 Authors

Abstract: We propose DiffCLIP, a novel vision-language model that extends the differential attention mechanism to CLIP architectures. Differential attention was originally developed for large language models to amplify relevant context while canceling out noisy information. In this work, we integrate this mechanism into CLIP's dual encoder (image and text) framework. With minimal additional parameters, DiffCLIP achieves superior performance on image-text understanding tasks. Across zero-shot classification, retrieval, and robustness benchmarks, DiffCLIP consistently outperforms baseline CLIP models. Notably, these gains come with negligible computational overhead, demonstrating that differential attention can significantly enhance multi-modal representations without sacrificing efficiency.

Submission Length: Regular submission (no more than 12 pages of main content)

Assigned Action Editor: ~Massimiliano_Mancini1

Submission Number: 4934