{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:XLACN233SBWALU6FDCSGLOAU2Y","short_pith_number":"pith:XLACN233","schema_version":"1.0","canonical_sha256":"bac026eb7b906c05d3c518a465b814d63330052af61b4cb26a700907b16b6da0","source":{"kind":"arxiv","id":"1909.03918","version":2},"attestation_state":"computed","paper":{"title":"Hierarchy Parsing for Image Captioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Tao Mei, Ting Yao, Yehao Li, Yingwei Pan","submitted_at":"2019-09-09T15:18:21Z","abstract_excerpt":"It is always well believed that parsing an image into constituent visual patterns would be helpful for understanding and representing an image. Nevertheless, there has not been evidence in support of the idea on describing an image with a natural-language utterance. In this paper, we introduce a new design to model a hierarchy from instance level (segmentation), region level (detection) to the whole image to delve into a thorough image understanding for captioning. Specifically, we present a HIerarchy Parsing (HIP) architecture that novelly integrates hierarchical structure into image encoder."},"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":"1909.03918","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-09-09T15:18:21Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f6c7958dc94f771b9836c900af5e57df3a320cfdf7b93a2040efc1d81c59ef33","abstract_canon_sha256":"ccd6cca1a0ecab7cf9ceb8807e8e2ca096ffbb88706f2957e976ffd69c345010"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:03:18.312693Z","signature_b64":"MihVVvNjt8bUEHSD7XqCxcPXFoYmcK9NPVlpYLC8OFW06oqgYpbI+4qYknrlPJ8qR87RoCadWJDNLbzLKbmQDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bac026eb7b906c05d3c518a465b814d63330052af61b4cb26a700907b16b6da0","last_reissued_at":"2026-07-05T00:03:18.311690Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:03:18.311690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hierarchy Parsing for Image Captioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Tao Mei, Ting Yao, Yehao Li, Yingwei Pan","submitted_at":"2019-09-09T15:18:21Z","abstract_excerpt":"It is always well believed that parsing an image into constituent visual patterns would be helpful for understanding and representing an image. Nevertheless, there has not been evidence in support of the idea on describing an image with a natural-language utterance. In this paper, we introduce a new design to model a hierarchy from instance level (segmentation), region level (detection) to the whole image to delve into a thorough image understanding for captioning. Specifically, we present a HIerarchy Parsing (HIP) architecture that novelly integrates hierarchical structure into image encoder."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.03918","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/1909.03918/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":"1909.03918","created_at":"2026-07-05T00:03:18.311749+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.03918v2","created_at":"2026-07-05T00:03:18.311749+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.03918","created_at":"2026-07-05T00:03:18.311749+00:00"},{"alias_kind":"pith_short_12","alias_value":"XLACN233SBWA","created_at":"2026-07-05T00:03:18.311749+00:00"},{"alias_kind":"pith_short_16","alias_value":"XLACN233SBWALU6F","created_at":"2026-07-05T00:03:18.311749+00:00"},{"alias_kind":"pith_short_8","alias_value":"XLACN233","created_at":"2026-07-05T00:03:18.311749+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/XLACN233SBWALU6FDCSGLOAU2Y","json":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y.json","graph_json":"https://pith.science/api/pith-number/XLACN233SBWALU6FDCSGLOAU2Y/graph.json","events_json":"https://pith.science/api/pith-number/XLACN233SBWALU6FDCSGLOAU2Y/events.json","paper":"https://pith.science/paper/XLACN233"},"agent_actions":{"view_html":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y","download_json":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y.json","view_paper":"https://pith.science/paper/XLACN233","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.03918&json=true","fetch_graph":"https://pith.science/api/pith-number/XLACN233SBWALU6FDCSGLOAU2Y/graph.json","fetch_events":"https://pith.science/api/pith-number/XLACN233SBWALU6FDCSGLOAU2Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y/action/storage_attestation","attest_author":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y/action/author_attestation","sign_citation":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y/action/citation_signature","submit_replication":"https://pith.science/pith/XLACN233SBWALU6FDCSGLOAU2Y/action/replication_record"}},"created_at":"2026-07-05T00:03:18.311749+00:00","updated_at":"2026-07-05T00:03:18.311749+00:00"}