{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:YI3DPXPKUTNTQLU7STMOI2PYQO","short_pith_number":"pith:YI3DPXPK","schema_version":"1.0","canonical_sha256":"c23637ddeaa4db382e9f94d8e469f883ab292666b6d744733b19e733202c97a3","source":{"kind":"arxiv","id":"2212.00281","version":2},"attestation_state":"computed","paper":{"title":"Localization vs. Semantics: Visual Representations in Unimodal and Multimodal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Alan Yuille, Benjamin Van Durme, Cihang Xie, Zhuowan Li","submitted_at":"2022-12-01T05:00:18Z","abstract_excerpt":"Despite the impressive advancements achieved through vision-and-language pretraining, it remains unclear whether this joint learning paradigm can help understand each individual modality. In this work, we conduct a comparative analysis of the visual representations in existing vision-and-language models and vision-only models by probing a broad range of tasks, aiming to assess the quality of the learned representations in a nuanced manner. Interestingly, our empirical observations suggest that vision-and-language models are better at label prediction tasks like object and attribute prediction,"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2212.00281","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-12-01T05:00:18Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"3f4748725d1bd37ff8fe4ce5f5e70d6f3412d5ebaf6886d41158ec37ff825099","abstract_canon_sha256":"88cee4681dc4373427406a2a3a2c412cc517c18e70de884ef3ca8928e37181f7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:38:49.854382Z","signature_b64":"lhGPJJk+Nf1KnEYnx2uCmm9zrqrW0UcXkJ6mhQii00c3Y6nghy0KRonOC4jtyOPJDbzOHzMMHYhGB97g4gwcAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c23637ddeaa4db382e9f94d8e469f883ab292666b6d744733b19e733202c97a3","last_reissued_at":"2026-07-05T07:38:49.853901Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:38:49.853901Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Localization vs. Semantics: Visual Representations in Unimodal and Multimodal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Alan Yuille, Benjamin Van Durme, Cihang Xie, Zhuowan Li","submitted_at":"2022-12-01T05:00:18Z","abstract_excerpt":"Despite the impressive advancements achieved through vision-and-language pretraining, it remains unclear whether this joint learning paradigm can help understand each individual modality. In this work, we conduct a comparative analysis of the visual representations in existing vision-and-language models and vision-only models by probing a broad range of tasks, aiming to assess the quality of the learned representations in a nuanced manner. Interestingly, our empirical observations suggest that vision-and-language models are better at label prediction tasks like object and attribute prediction,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.00281","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2212.00281/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2212.00281","created_at":"2026-07-05T07:38:49.853963+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.00281v2","created_at":"2026-07-05T07:38:49.853963+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.00281","created_at":"2026-07-05T07:38:49.853963+00:00"},{"alias_kind":"pith_short_12","alias_value":"YI3DPXPKUTNT","created_at":"2026-07-05T07:38:49.853963+00:00"},{"alias_kind":"pith_short_16","alias_value":"YI3DPXPKUTNTQLU7","created_at":"2026-07-05T07:38:49.853963+00:00"},{"alias_kind":"pith_short_8","alias_value":"YI3DPXPK","created_at":"2026-07-05T07:38:49.853963+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO","json":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO.json","graph_json":"https://pith.science/api/pith-number/YI3DPXPKUTNTQLU7STMOI2PYQO/graph.json","events_json":"https://pith.science/api/pith-number/YI3DPXPKUTNTQLU7STMOI2PYQO/events.json","paper":"https://pith.science/paper/YI3DPXPK"},"agent_actions":{"view_html":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO","download_json":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO.json","view_paper":"https://pith.science/paper/YI3DPXPK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.00281&json=true","fetch_graph":"https://pith.science/api/pith-number/YI3DPXPKUTNTQLU7STMOI2PYQO/graph.json","fetch_events":"https://pith.science/api/pith-number/YI3DPXPKUTNTQLU7STMOI2PYQO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO/action/storage_attestation","attest_author":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO/action/author_attestation","sign_citation":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO/action/citation_signature","submit_replication":"https://pith.science/pith/YI3DPXPKUTNTQLU7STMOI2PYQO/action/replication_record"}},"created_at":"2026-07-05T07:38:49.853963+00:00","updated_at":"2026-07-05T07:38:49.853963+00:00"}