{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/calip-zero-shot-enhancement-of-clip-with","title":"CALIP: Zero-Shot Enhancement of CLIP with Parameter-free Attention","arxiv_id":"2209.14169","date":"2022-09-28","proceeding":null,"authors":["Ziyu Guo","Renrui Zhang","Longtian Qiu","Xianzheng Ma","Xupeng Miao","Xuming He","Bin Cui"],"abstract":"Contrastive Language-Image Pre-training (CLIP) has been shown to learn visual representations with great transferability, which achieves promising accuracy for zero-shot classification. To further improve its downstream performance, existing works propose additional learnable modules upon CLIP and fine-tune them by few-shot training sets. However, the resulting extra training cost and data requirement severely hinder the efficiency for model deployment and knowledge transfer. In this paper, we introduce a free-lunch enhancement method, CALIP, to boost CLIP's zero-shot performance via a parameter-free Attention module. Specifically, we guide visual and textual representations to interact with each other and explore cross-modal informative features via attention. As the pre-training has largely reduced the embedding distances between two modalities, we discard all learnable parameters in the attention and bidirectionally update the multi-modal features, enabling the whole process to be parameter-free and training-free. In this way, the images are blended with textual-aware signals and the text representations become visual-guided for better adaptive zero-shot alignment. We evaluate CALIP on various benchmarks of 14 datasets for both 2D image and 3D point cloud few-shot classification, showing consistent zero-shot performance improvement over CLIP. Based on that, we further insert a small number of linear layers in CALIP's attention module and verify our robustness under the few-shot settings, which also achieves leading performance compared to existing methods. Those extensive experiments demonstrate the superiority of our approach for efficient enhancement of CLIP.","url_abs":"https://arxiv.org/abs/2209.14169v2","url_pdf":"https://arxiv.org/pdf/2209.14169v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"calip-zero-shot-enhancement-of-clip-with","repo_url":"https://github.com/ziyuguo99/calip","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"training-free-3d-point-cloud-classification","task_name":"Training-free 3D Point Cloud Classification"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"},{"task_slug":null,"task_name":"zero-shot-classification"}],"methods":[{"method_slug":"clip","method_name":"CLIP"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/training-free-3d-point-cloud-classification","task":"Training-free 3D Point Cloud Classification","dataset":"ModelNet40","model":"CALIP","rank_in_archive_order":6,"of":7,"metrics":{"Accuracy (%)":"21.5","Need 3D Data?":"No"},"uses_additional_data":true},{"leaderboard":"/sota/training-free-3d-point-cloud-classification-1","task":"Training-free 3D Point Cloud Classification","dataset":"ScanObjectNN","model":"CALIP","rank_in_archive_order":5,"of":6,"metrics":{"Accuracy (%)":"16.9","Need 3D Data?":"No"},"uses_additional_data":true}],"syntology":{"syntology_url":"https://syntology.ai/paper/2209.14169","atlas_url":"https://app.syntology.ai/?focus=2209.14169","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}