{"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/structure-clip-enhance-multi-modal-language","title":"Structure-CLIP: Towards Scene Graph Knowledge to Enhance Multi-modal Structured Representations","arxiv_id":"2305.06152","date":"2023-05-06","proceeding":null,"authors":["Yufeng Huang","Jiji Tang","Zhuo Chen","Rongsheng Zhang","Xinfeng Zhang","WeiJie Chen","Zeng Zhao","Zhou Zhao","Tangjie Lv","Zhipeng Hu","Wen Zhang"],"abstract":"Large-scale vision-language pre-training has achieved significant performance in multi-modal understanding and generation tasks. However, existing methods often perform poorly on image-text matching tasks that require structured representations, i.e., representations of objects, attributes, and relations. As illustrated in Fig.~reffig:case (a), the models cannot make a distinction between ``An astronaut rides a horse\" and ``A horse rides an astronaut\". This is because they fail to fully leverage structured knowledge when learning representations in multi-modal scenarios. In this paper, we present an end-to-end framework Structure-CLIP, which integrates Scene Graph Knowledge (SGK) to enhance multi-modal structured representations. Firstly, we use scene graphs to guide the construction of semantic negative examples, which results in an increased emphasis on learning structured representations. Moreover, a Knowledge-Enhance Encoder (KEE) is proposed to leverage SGK as input to further enhance structured representations. To verify the effectiveness of the proposed framework, we pre-train our model with the aforementioned approaches and conduct experiments on downstream tasks. Experimental results demonstrate that Structure-CLIP achieves state-of-the-art (SOTA) performance on VG-Attribution and VG-Relation datasets, with 12.5% and 4.1% ahead of the multi-modal SOTA model respectively. Meanwhile, the results on MSCOCO indicate that Structure-CLIP significantly enhances the structured representations while maintaining the ability of general representations. Our code is available at https://github.com/zjukg/Structure-CLIP.","url_abs":"https://arxiv.org/abs/2305.06152v3","url_pdf":"https://arxiv.org/pdf/2305.06152v3.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":"structure-clip-enhance-multi-modal-language","repo_url":"https://github.com/zjukg/structure-clip","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"structure-clip-enhance-multi-modal-language","repo_url":"https://github.com/zjukg/DUET","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"structure-clip-enhance-multi-modal-language","repo_url":"https://github.com/zjukg/trustuqa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-text-matching","task_name":"Image-text matching"},{"task_slug":"text-matching","task_name":"Text Matching"}],"methods":[{"method_slug":"fail","method_name":"fail"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2305.06152","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2305.06152"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+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/zjukg/trustuqa","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/zjukg/DUET","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/zjukg/structure-clip","reach":{"status":"ok"}}],"summary":{"unverified":9},"by_repo_kind":{"listed":{"samples":9,"ran":0,"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":0,"samples":[{"code_sha256_prefix":"499944ccee88c36f","entry":"L2_normalize","repo":"zjukg/DUET","repo_kind":"listed","path":"model/visual_utils.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/visual_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"499944ccee88c36f"}},{"code_sha256_prefix":"a2b367aa9ac6d4c8","entry":"calibrated_stacking","repo":"zjukg/DUET","repo_kind":"listed","path":"model/main_utils.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/main_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a2b367aa9ac6d4c8"}},{"code_sha256_prefix":"cad018a9a75d5d4b","entry":"compute_sim","repo":"zjukg/DUET","repo_kind":"listed","path":"model/util.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cad018a9a75d5d4b"}},{"code_sha256_prefix":"783a0907b8c9734d","entry":"cosine_similarity","repo":"zjukg/DUET","repo_kind":"listed","path":"model/util.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"783a0907b8c9734d"}},{"code_sha256_prefix":"4db905959447729d","entry":"load_tf_weights_in_bert","repo":"zjukg/DUET","repo_kind":"listed","path":"model/swin_modeling_bert.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/swin_modeling_bert.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4db905959447729d"}},{"code_sha256_prefix":"0dc3270313a8304e","entry":"load_tf_weights_in_lxmert","repo":"zjukg/DUET","repo_kind":"listed","path":"model/modeling_lxmert.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/modeling_lxmert.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0dc3270313a8304e"}},{"code_sha256_prefix":"a8962fd16b659954","entry":"map_label","repo":"zjukg/DUET","repo_kind":"listed","path":"model/visual_utils.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/visual_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a8962fd16b659954"}},{"code_sha256_prefix":"4f51d40e5e4d5244","entry":"randomMask","repo":"zjukg/DUET","repo_kind":"listed","path":"model/util.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4f51d40e5e4d5244"}},{"code_sha256_prefix":"e9f8277102262110","entry":"sort_acc","repo":"zjukg/DUET","repo_kind":"listed","path":"model/visual_utils.py","file_url":"https://github.com/zjukg/DUET/blob/HEAD/model/visual_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e9f8277102262110"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}