{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:BN3II3BO7A4LWTYI6LECRIICEZ","short_pith_number":"pith:BN3II3BO","schema_version":"1.0","canonical_sha256":"0b76846c2ef838bb4f08f2c828a1022675d0b0047a31962a5d098047b184b438","source":{"kind":"arxiv","id":"2303.17590","version":2},"attestation_state":"computed","paper":{"title":"Going Beyond Nouns With Vision & Language Models Using Synthetic Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Aude Oliva, Donghyun Kim, G\\\"ul Varol, James Seale Smith, Khaled Shehada, Leonid Karlinsky, Paola Cascante-Bonilla, Rameswar Panda, Rogerio Feris, Sivan Doveh, Vicente Ordonez","submitted_at":"2023-03-30T17:57:43Z","abstract_excerpt":"Large-scale pre-trained Vision & Language (VL) models have shown remarkable performance in many applications, enabling replacing a fixed set of supported classes with zero-shot open vocabulary reasoning over (almost arbitrary) natural language prompts. However, recent works have uncovered a fundamental weakness of these models. For example, their difficulty to understand Visual Language Concepts (VLC) that go 'beyond nouns' such as the meaning of non-object words (e.g., attributes, actions, relations, states, etc.), or difficulty in performing compositional reasoning such as understanding the "},"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":"2303.17590","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-30T17:57:43Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"859deb6746a4728204b355c9c991309f05600ddf69bdb046d4c1c0faaac815c3","abstract_canon_sha256":"b9212b3b5109f8f4a947aae8dc04c75b00a4b7528e5659b6f6a59665f02eaaf2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:46:02.081209Z","signature_b64":"cIdbHUyfdMrSYkEHpZWJH8Etel8yCsQkq+JyrAdeGXGvRx13XdK5vMgb+TCXqIr2HXTkfkyWTqoZM7DFttUvBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b76846c2ef838bb4f08f2c828a1022675d0b0047a31962a5d098047b184b438","last_reissued_at":"2026-07-05T06:46:02.080739Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:46:02.080739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Going Beyond Nouns With Vision & Language Models Using Synthetic Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Aude Oliva, Donghyun Kim, G\\\"ul Varol, James Seale Smith, Khaled Shehada, Leonid Karlinsky, Paola Cascante-Bonilla, Rameswar Panda, Rogerio Feris, Sivan Doveh, Vicente Ordonez","submitted_at":"2023-03-30T17:57:43Z","abstract_excerpt":"Large-scale pre-trained Vision & Language (VL) models have shown remarkable performance in many applications, enabling replacing a fixed set of supported classes with zero-shot open vocabulary reasoning over (almost arbitrary) natural language prompts. However, recent works have uncovered a fundamental weakness of these models. For example, their difficulty to understand Visual Language Concepts (VLC) that go 'beyond nouns' such as the meaning of non-object words (e.g., attributes, actions, relations, states, etc.), or difficulty in performing compositional reasoning such as understanding the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.17590","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/2303.17590/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":"2303.17590","created_at":"2026-07-05T06:46:02.080808+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.17590v2","created_at":"2026-07-05T06:46:02.080808+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.17590","created_at":"2026-07-05T06:46:02.080808+00:00"},{"alias_kind":"pith_short_12","alias_value":"BN3II3BO7A4L","created_at":"2026-07-05T06:46:02.080808+00:00"},{"alias_kind":"pith_short_16","alias_value":"BN3II3BO7A4LWTYI","created_at":"2026-07-05T06:46:02.080808+00:00"},{"alias_kind":"pith_short_8","alias_value":"BN3II3BO","created_at":"2026-07-05T06:46:02.080808+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/BN3II3BO7A4LWTYI6LECRIICEZ","json":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ.json","graph_json":"https://pith.science/api/pith-number/BN3II3BO7A4LWTYI6LECRIICEZ/graph.json","events_json":"https://pith.science/api/pith-number/BN3II3BO7A4LWTYI6LECRIICEZ/events.json","paper":"https://pith.science/paper/BN3II3BO"},"agent_actions":{"view_html":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ","download_json":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ.json","view_paper":"https://pith.science/paper/BN3II3BO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.17590&json=true","fetch_graph":"https://pith.science/api/pith-number/BN3II3BO7A4LWTYI6LECRIICEZ/graph.json","fetch_events":"https://pith.science/api/pith-number/BN3II3BO7A4LWTYI6LECRIICEZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ/action/storage_attestation","attest_author":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ/action/author_attestation","sign_citation":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ/action/citation_signature","submit_replication":"https://pith.science/pith/BN3II3BO7A4LWTYI6LECRIICEZ/action/replication_record"}},"created_at":"2026-07-05T06:46:02.080808+00:00","updated_at":"2026-07-05T06:46:02.080808+00:00"}