{"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/learning-visually-grounded-semantics-from","title":"Learning Visually-Grounded Semantics from Contrastive Adversarial Samples","arxiv_id":"1806.10348","date":"2018-06-27","proceeding":"COLING 2018 8","authors":["Haoyue Shi","Jiayuan Mao","Tete Xiao","Yuning Jiang","Jian Sun"],"abstract":"We study the problem of grounding distributional representations of texts on\nthe visual domain, namely visual-semantic embeddings (VSE for short). Begin\nwith an insightful adversarial attack on VSE embeddings, we show the limitation\nof current frameworks and image-text datasets (e.g., MS-COCO) both\nquantitatively and qualitatively. The large gap between the number of possible\nconstitutions of real-world semantics and the size of parallel data, to a large\nextent, restricts the model to establish the link between textual semantics and\nvisual concepts. We alleviate this problem by augmenting the MS-COCO image\ncaptioning datasets with textual contrastive adversarial samples. These samples\nare synthesized using linguistic rules and the WordNet knowledge base. The\nconstruction procedure is both syntax- and semantics-aware. The samples enforce\nthe model to ground learned embeddings to concrete concepts within the image.\nThis simple but powerful technique brings a noticeable improvement over the\nbaselines on a diverse set of downstream tasks, in addition to defending\nknown-type adversarial attacks. We release the codes at\nhttps://github.com/ExplorerFreda/VSE-C.","url_abs":"http://arxiv.org/abs/1806.10348v1","url_pdf":"http://arxiv.org/pdf/1806.10348v1.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":"learning-visually-grounded-semantics-from","repo_url":"https://github.com/ExplorerFreda/VSE-C","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"adversarial-attack","task_name":"Adversarial Attack"},{"task_slug":"image-captioning","task_name":"Image Captioning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1806.10348","atlas_url":"https://app.syntology.ai/?focus=1806.10348","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1806.10348"}},"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/ExplorerFreda/VSE-C","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"unverified":13},"by_repo_kind":{"official":{"samples":15,"ran":2,"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":"9b0a787b92a87023","entry":"cosine_sim","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/model.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/model.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9b0a787b92a87023"}},{"code_sha256_prefix":"ef05dda21c1cc726","entry":"from_txt","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/vocab.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/vocab.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ef05dda21c1cc726"}},{"code_sha256_prefix":"16c5b66114f6387f","entry":"EncoderImage","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/model.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/model.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":"16c5b66114f6387f"}},{"code_sha256_prefix":"e143fe1425b0601d","entry":"English","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"adversarial_attack/noun.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/adversarial_attack/noun.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":"e143fe1425b0601d"}},{"code_sha256_prefix":"472607bbb393c46f","entry":"English","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"adversarial_attack/relation.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/adversarial_attack/relation.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":"472607bbb393c46f"}},{"code_sha256_prefix":"6961e1859e3c7fe9","entry":"collate_fn","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/data.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/data.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":"6961e1859e3c7fe9"}},{"code_sha256_prefix":"1a58cc01e62897f2","entry":"collate_fn_train_text","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/data.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/data.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":"1a58cc01e62897f2"}},{"code_sha256_prefix":"d113e4ae72bec1ef","entry":"detect_noun_chunks","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"adversarial_attack/relation.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/adversarial_attack/relation.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":"d113e4ae72bec1ef"}},{"code_sha256_prefix":"5eef3e620d10d47b","entry":"encode_data","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/evaluation.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/evaluation.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":"5eef3e620d10d47b"}},{"code_sha256_prefix":"4d60d6b4f58a577f","entry":"from_flickr_json","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/vocab.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/vocab.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":"4d60d6b4f58a577f"}},{"code_sha256_prefix":"b61084903f7e3069","entry":"get_paths","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/data.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/data.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":"b61084903f7e3069"}},{"code_sha256_prefix":"7a8088198ac7ad87","entry":"isalpha","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"evaluation/completion/completion_datamaker.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/evaluation/completion/completion_datamaker.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":"7a8088198ac7ad87"}},{"code_sha256_prefix":"1ff2d83dc35a67ce","entry":"l2norm","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"VSE_C/model.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/VSE_C/model.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":"1ff2d83dc35a67ce"}},{"code_sha256_prefix":"373106f951241575","entry":"match","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"adversarial_attack/relation.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/adversarial_attack/relation.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":"373106f951241575"}},{"code_sha256_prefix":"4839bd92d6cdf63b","entry":"valid","repo":"ExplorerFreda/VSE-C","repo_kind":"official","path":"adversarial_attack/noun.py","file_url":"https://github.com/ExplorerFreda/VSE-C/blob/HEAD/adversarial_attack/noun.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":"4839bd92d6cdf63b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}