{"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/caila-concept-aware-intra-layer-adapters-for","title":"CAILA: Concept-Aware Intra-Layer Adapters for Compositional Zero-Shot Learning","arxiv_id":"2305.16681","date":"2023-05-26","proceeding":null,"authors":["Zhaoheng Zheng","Haidong Zhu","Ram Nevatia"],"abstract":"In this paper, we study the problem of Compositional Zero-Shot Learning (CZSL), which is to recognize novel attribute-object combinations with pre-existing concepts. Recent researchers focus on applying large-scale Vision-Language Pre-trained (VLP) models like CLIP with strong generalization ability. However, these methods treat the pre-trained model as a black box and focus on pre- and post-CLIP operations, which do not inherently mine the semantic concept between the layers inside CLIP. We propose to dive deep into the architecture and insert adapters, a parameter-efficient technique proven to be effective among large language models, into each CLIP encoder layer. We further equip adapters with concept awareness so that concept-specific features of \"object\", \"attribute\", and \"composition\" can be extracted. We assess our method on four popular CZSL datasets, MIT-States, C-GQA, UT-Zappos, and VAW-CZSL, which shows state-of-the-art performance compared to existing methods on all of them.","url_abs":"https://arxiv.org/abs/2305.16681v2","url_pdf":"https://arxiv.org/pdf/2305.16681v2.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":"caila-concept-aware-intra-layer-adapters-for","repo_url":"https://github.com/zhaohengz/caila","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"caila-concept-aware-intra-layer-adapters-for","repo_url":"https://github.com/zhaohengz/llamp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"compositional-zero-shot-learning","task_name":"Compositional Zero-Shot Learning"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"}],"methods":[{"method_slug":"clip","method_name":"CLIP"},{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/compositional-zero-shot-learning-on-mit-3","task":"Compositional Zero-Shot Learning","dataset":"MIT-States, generalized split","model":"CAILA","rank_in_archive_order":1,"of":2,"metrics":{"H-Mean":"39.9","Seen accuracy":"51.0","Test AUC top 1":"23.4","Test AUC top 2":"-","Test AUC top 3":"-","Unseen accuracy":"53.9","Val AUC top 1":"-","Val AUC top 2":"-","Val AUC top 3":"-"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2305.16681","atlas_url":"https://app.syntology.ai/?focus=2305.16681","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}