{"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/leveraging-the-invariant-side-of-generative","title":"Leveraging the Invariant Side of Generative Zero-Shot Learning","arxiv_id":"1904.04092","date":"2019-04-08","proceeding":"CVPR 2019 6","authors":["Jingjing Li","Mengmeng Jin","Ke Lu","Zhengming Ding","Lei Zhu","Zi Huang"],"abstract":"Conventional zero-shot learning (ZSL) methods generally learn an embedding,\ne.g., visual-semantic mapping, to handle the unseen visual samples via an\nindirect manner. In this paper, we take the advantage of generative adversarial\nnetworks (GANs) and propose a novel method, named leveraging invariant side GAN\n(LisGAN), which can directly generate the unseen features from random noises\nwhich are conditioned by the semantic descriptions. Specifically, we train a\nconditional Wasserstein GANs in which the generator synthesizes fake unseen\nfeatures from noises and the discriminator distinguishes the fake from real via\na minimax game. Considering that one semantic description can correspond to\nvarious synthesized visual samples, and the semantic description, figuratively,\nis the soul of the generated features, we introduce soul samples as the\ninvariant side of generative zero-shot learning in this paper. A soul sample is\nthe meta-representation of one class. It visualizes the most\nsemantically-meaningful aspects of each sample in the same category. We\nregularize that each generated sample (the varying side of generative ZSL)\nshould be close to at least one soul sample (the invariant side) which has the\nsame class label with it. At the zero-shot recognition stage, we propose to use\ntwo classifiers, which are deployed in a cascade way, to achieve a\ncoarse-to-fine result. Experiments on five popular benchmarks verify that our\nproposed approach can outperform state-of-the-art methods with significant\nimprovements.","url_abs":"http://arxiv.org/abs/1904.04092v1","url_pdf":"http://arxiv.org/pdf/1904.04092v1.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":"leveraging-the-invariant-side-of-generative","repo_url":"https://github.com/lijin118/LisGAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"generalized-zero-shot-learning","task_name":"Generalized Zero-Shot Learning"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/generalized-zero-shot-learning-on-sun","task":"Generalized Zero-Shot Learning","dataset":"SUN Attribute","model":"LisGAN","rank_in_archive_order":5,"of":9,"metrics":{"Harmonic mean":"40.2"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-learning-on-cub-200-2011","task":"Zero-Shot Learning","dataset":"CUB-200-2011","model":"LisGAN","rank_in_archive_order":10,"of":14,"metrics":{"average top-1 classification accuracy":"58.8"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-learning-on-sun-attribute","task":"Zero-Shot Learning","dataset":"SUN Attribute","model":"LisGAN","rank_in_archive_order":6,"of":9,"metrics":{"average top-1 classification accuracy":"61.7"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1904.04092","atlas_url":"https://app.syntology.ai/?focus=1904.04092","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1904.04092"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lijin118/LisGAN","reach":null}],"summary":{"ran_violates":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"7ac965f971d0020b","entry":"map_label","repo":"lijin118/LisGAN","repo_kind":"official","path":"lisgan.py","file_url":"https://github.com/lijin118/LisGAN/blob/HEAD/lisgan.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7ac965f971d0020b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}