{"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/duet-cross-modal-semantic-grounding-for","title":"DUET: Cross-modal Semantic Grounding for Contrastive Zero-shot Learning","arxiv_id":"2207.01328","date":"2022-07-04","proceeding":null,"authors":["Zhuo Chen","Yufeng Huang","Jiaoyan Chen","Yuxia Geng","Wen Zhang","Yin Fang","Jeff Z. Pan","Huajun Chen"],"abstract":"Zero-shot learning (ZSL) aims to predict unseen classes whose samples have never appeared during training. One of the most effective and widely used semantic information for zero-shot image classification are attributes which are annotations for class-level visual characteristics. However, the current methods often fail to discriminate those subtle visual distinctions between images due to not only the shortage of fine-grained annotations, but also the attribute imbalance and co-occurrence. In this paper, we present a transformer-based end-to-end ZSL method named DUET, which integrates latent semantic knowledge from the pre-trained language models (PLMs) via a self-supervised multi-modal learning paradigm. Specifically, we (1) developed a cross-modal semantic grounding network to investigate the model's capability of disentangling semantic attributes from the images; (2) applied an attribute-level contrastive learning strategy to further enhance the model's discrimination on fine-grained visual characteristics against the attribute co-occurrence and imbalance; (3) proposed a multi-task learning policy for considering multi-model objectives. We find that our DUET can achieve state-of-the-art performance on three standard ZSL benchmarks and a knowledge graph equipped ZSL benchmark. Its components are effective and its predictions are interpretable.","url_abs":"https://arxiv.org/abs/2207.01328v4","url_pdf":"https://arxiv.org/pdf/2207.01328v4.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":"duet-cross-modal-semantic-grounding-for","repo_url":"https://github.com/zjukg/DUET","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"duet-cross-modal-semantic-grounding-for","repo_url":"https://github.com/zjukg/structure-clip","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"contrastive-learning","task_name":"Contrastive Learning"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"zero-shot-image-classification","task_name":"Zero-Shot Image Classification"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"contrastive-learning","method_name":"Contrastive Learning"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/zero-shot-learning-on-awa2","task":"Zero-Shot Learning","dataset":"AwA2","model":"DUET (Ours)","rank_in_archive_order":4,"of":4,"metrics":{"Accuracy Seen":"84.7","Accuracy Unseen":"63.7","H":"72.7","average top-1 classification accuracy":"69.9"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-learning-on-cub-200-2011","task":"Zero-Shot Learning","dataset":"CUB-200-2011","model":"DUET","rank_in_archive_order":2,"of":14,"metrics":{"Accuracy Seen":"72.8","Accuracy Unseen":"62.9","H":"67.5","average top-1 classification accuracy":"72.3"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-learning-on-sun-attribute","task":"Zero-Shot Learning","dataset":"SUN Attribute","model":"DUET (Ours)","rank_in_archive_order":5,"of":9,"metrics":{"Accuracy Seen":"45.8","Accuracy Unseen":"45.7","H":"45.8","average top-1 classification accuracy":"64.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2207.01328","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}