{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:5HTIQIYAC7H5SJI7HCGXG3AKGF","short_pith_number":"pith:5HTIQIYA","schema_version":"1.0","canonical_sha256":"e9e688230017cfd9251f388d736c0a317ed5db691f363820d0c65b1698774559","source":{"kind":"arxiv","id":"2203.14260","version":3},"attestation_state":"computed","paper":{"title":"Unsupervised Vision-Language Parsing: Seamlessly Bridging Visual Scene Graphs with Language Structures via Dependency Relationships","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Chao Lou, Wenjuan Han, Yuhuan Lin, Zilong Zheng","submitted_at":"2022-03-27T09:51:34Z","abstract_excerpt":"Understanding realistic visual scene images together with language descriptions is a fundamental task towards generic visual understanding. Previous works have shown compelling comprehensive results by building hierarchical structures for visual scenes (e.g., scene graphs) and natural languages (e.g., dependency trees), individually. However, how to construct a joint vision-language (VL) structure has barely been investigated. More challenging but worthwhile, we introduce a new task that targets on inducing such a joint VL structure in an unsupervised manner. Our goal is to bridge the visual s"},"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":"2203.14260","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2022-03-27T09:51:34Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"d925671f169734afbfcd9a1ead1eaee7530276acf188587a8427632b87945545","abstract_canon_sha256":"682dcc8463f1dff39de9b8fb07bd38cfe7da0c70f12d18120e9c8174d2be4255"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:28:10.839146Z","signature_b64":"91O6trq2ruHnEEXrAiZpR4ZYoVaEFSR6D/GgTMyVFe7RtnEV4J7mUCgAp5XbYBb8rk06w39djZ8b+JKFI6M3DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e9e688230017cfd9251f388d736c0a317ed5db691f363820d0c65b1698774559","last_reissued_at":"2026-07-05T04:28:10.838641Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:28:10.838641Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unsupervised Vision-Language Parsing: Seamlessly Bridging Visual Scene Graphs with Language Structures via Dependency Relationships","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Chao Lou, Wenjuan Han, Yuhuan Lin, Zilong Zheng","submitted_at":"2022-03-27T09:51:34Z","abstract_excerpt":"Understanding realistic visual scene images together with language descriptions is a fundamental task towards generic visual understanding. Previous works have shown compelling comprehensive results by building hierarchical structures for visual scenes (e.g., scene graphs) and natural languages (e.g., dependency trees), individually. However, how to construct a joint vision-language (VL) structure has barely been investigated. More challenging but worthwhile, we introduce a new task that targets on inducing such a joint VL structure in an unsupervised manner. Our goal is to bridge the visual s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.14260","kind":"arxiv","version":3},"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/2203.14260/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":"2203.14260","created_at":"2026-07-05T04:28:10.838703+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.14260v3","created_at":"2026-07-05T04:28:10.838703+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.14260","created_at":"2026-07-05T04:28:10.838703+00:00"},{"alias_kind":"pith_short_12","alias_value":"5HTIQIYAC7H5","created_at":"2026-07-05T04:28:10.838703+00:00"},{"alias_kind":"pith_short_16","alias_value":"5HTIQIYAC7H5SJI7","created_at":"2026-07-05T04:28:10.838703+00:00"},{"alias_kind":"pith_short_8","alias_value":"5HTIQIYA","created_at":"2026-07-05T04:28:10.838703+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/5HTIQIYAC7H5SJI7HCGXG3AKGF","json":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF.json","graph_json":"https://pith.science/api/pith-number/5HTIQIYAC7H5SJI7HCGXG3AKGF/graph.json","events_json":"https://pith.science/api/pith-number/5HTIQIYAC7H5SJI7HCGXG3AKGF/events.json","paper":"https://pith.science/paper/5HTIQIYA"},"agent_actions":{"view_html":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF","download_json":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF.json","view_paper":"https://pith.science/paper/5HTIQIYA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.14260&json=true","fetch_graph":"https://pith.science/api/pith-number/5HTIQIYAC7H5SJI7HCGXG3AKGF/graph.json","fetch_events":"https://pith.science/api/pith-number/5HTIQIYAC7H5SJI7HCGXG3AKGF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF/action/storage_attestation","attest_author":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF/action/author_attestation","sign_citation":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF/action/citation_signature","submit_replication":"https://pith.science/pith/5HTIQIYAC7H5SJI7HCGXG3AKGF/action/replication_record"}},"created_at":"2026-07-05T04:28:10.838703+00:00","updated_at":"2026-07-05T04:28:10.838703+00:00"}