{"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/crepe-can-vision-language-foundation-models","title":"CREPE: Can Vision-Language Foundation Models Reason Compositionally?","arxiv_id":"2212.07796","date":"2022-12-13","proceeding":"CVPR 2023 1","authors":["Zixian Ma","Jerry Hong","Mustafa Omer Gul","Mona Gandhi","Irena Gao","Ranjay Krishna"],"abstract":"A fundamental characteristic common to both human vision and natural language is their compositional nature. Yet, despite the performance gains contributed by large vision and language pretraining, we find that: across 7 architectures trained with 4 algorithms on massive datasets, they struggle at compositionality. To arrive at this conclusion, we introduce a new compositionality evaluation benchmark, CREPE, which measures two important aspects of compositionality identified by cognitive science literature: systematicity and productivity. To measure systematicity, CREPE consists of a test dataset containing over $370K$ image-text pairs and three different seen-unseen splits. The three splits are designed to test models trained on three popular training datasets: CC-12M, YFCC-15M, and LAION-400M. We also generate $325K$, $316K$, and $309K$ hard negative captions for a subset of the pairs. To test productivity, CREPE contains $17K$ image-text pairs with nine different complexities plus $183K$ hard negative captions with atomic, swapping and negation foils. The datasets are generated by repurposing the Visual Genome scene graphs and region descriptions and applying handcrafted templates and GPT-3. For systematicity, we find that model performance decreases consistently when novel compositions dominate the retrieval set, with Recall@1 dropping by up to $12\\%$. For productivity, models' retrieval success decays as complexity increases, frequently nearing random chance at high complexity. These results hold regardless of model and training dataset size.","url_abs":"https://arxiv.org/abs/2212.07796v3","url_pdf":"https://arxiv.org/pdf/2212.07796v3.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":"crepe-can-vision-language-foundation-models","repo_url":"https://github.com/raivnlab/crepe","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"negation","task_name":"Negation"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-3","method_name":"GPT-3"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"test","method_name":"Test"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[{"slug":"crepe-vision-language","name":"CREPE (Compositional REPresentation Evaluation)","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"ViT-L-14 (LAION400M)","rank_in_archive_order":1,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"39.44","Recall@1 (HN-Atom + HN-Comp, UC)":"33.81","Recall@1 (HN-Atom, UC)":"47.86","Recall@1 (HN-Comp, UC)":"60.78"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"ViT-B-16+240 (LAION400M)","rank_in_archive_order":2,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"37.32","Recall@1 (HN-Atom + HN-Comp, UC)":"32.26","Recall@1 (HN-Atom, UC)":"46.53","Recall@1 (HN-Comp, UC)":"60.19"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"ViT-B-16 (LAION400M)","rank_in_archive_order":3,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"37.01","Recall@1 (HN-Atom + HN-Comp, UC)":"30.81","Recall@1 (HN-Atom, UC)":"44.93","Recall@1 (HN-Comp, UC)":"59.00"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"ViT-B-32 (LAION400M)","rank_in_archive_order":4,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"34.28","Recall@1 (HN-Atom + HN-Comp, UC)":"28.00","Recall@1 (HN-Atom, UC)":"42.75","Recall@1 (HN-Comp, UC)":"54.80"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"RN50 (YFCC15M)","rank_in_archive_order":5,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"23.38","Recall@1 (HN-Atom + HN-Comp, UC)":"20.08","Recall@1 (HN-Atom, UC)":"39.85","Recall@1 (HN-Comp, UC)":"39.83"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"RN50 (CC12M)","rank_in_archive_order":6,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"23.26","Recall@1 (HN-Atom + HN-Comp, UC)":"19.96","Recall@1 (HN-Atom, UC)":"34.88","Recall@1 (HN-Comp, UC)":"45.27"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"RN101 (YFCC15M)","rank_in_archive_order":7,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"22.74","Recall@1 (HN-Atom + HN-Comp, UC)":"20.50","Recall@1 (HN-Atom, UC)":"39.50","Recall@1 (HN-Comp, UC)":"39.56"},"uses_additional_data":false},{"leaderboard":"/sota/image-retrieval-on-crepe-vision-language","task":"Image Retrieval","dataset":"CREPE (Compositional REPresentation Evaluation)","model":"Random","rank_in_archive_order":8,"of":22,"metrics":{"Recall@1 (HN-Atom + HN-Comp, SC)":"9.09","Recall@1 (HN-Atom + HN-Comp, UC)":"9.09","Recall@1 (HN-Atom, UC)":"20.00","Recall@1 (HN-Comp, UC)":"14.29"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2212.07796","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2212.07796"}},"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/raivnlab/crepe","reach":null}],"summary":{"ran_draft_wrong":2,"ran_violates":1},"by_repo_kind":{"official":{"samples":3,"ran":3,"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":3,"samples":[{"code_sha256_prefix":"ac38eb0d52769ad0","entry":"get_metrics","repo":"raivnlab/crepe","repo_kind":"official","path":"crepe_eval_utils.py","file_url":"https://github.com/raivnlab/crepe/blob/HEAD/crepe_eval_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ac38eb0d52769ad0"}},{"code_sha256_prefix":"30948506734a8c2b","entry":"get_one2many_metrics","repo":"raivnlab/crepe","repo_kind":"official","path":"crepe_eval_utils.py","file_url":"https://github.com/raivnlab/crepe/blob/HEAD/crepe_eval_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"30948506734a8c2b"}},{"code_sha256_prefix":"50cb4e8382ec379d","entry":"get_one2many_rank","repo":"raivnlab/crepe","repo_kind":"official","path":"crepe_eval_utils.py","file_url":"https://github.com/raivnlab/crepe/blob/HEAD/crepe_eval_utils.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":"50cb4e8382ec379d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}