{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:4BYGN3IMBGAABW75FFGTB6DUGS","short_pith_number":"pith:4BYGN3IM","schema_version":"1.0","canonical_sha256":"e07066ed0c098000dbfd294d30f87434bedd077da0250d6601fad2a8491dc8e7","source":{"kind":"arxiv","id":"1905.02188","version":3},"attestation_state":"computed","paper":{"title":"CARAFE: Content-Aware ReAssembly of FEatures","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Change Loy, Dahua Lin, Jiaqi Wang, Kai Chen, Rui Xu, Ziwei Liu","submitted_at":"2019-05-06T17:58:06Z","abstract_excerpt":"Feature upsampling is a key operation in a number of modern convolutional network architectures, e.g. feature pyramids. Its design is critical for dense prediction tasks such as object detection and semantic/instance segmentation. In this work, we propose Content-Aware ReAssembly of FEatures (CARAFE), a universal, lightweight and highly effective operator to fulfill this goal. CARAFE has several appealing properties: (1) Large field of view. Unlike previous works (e.g. bilinear interpolation) that only exploit sub-pixel neighborhood, CARAFE can aggregate contextual information within a large r"},"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":"1905.02188","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-05-06T17:58:06Z","cross_cats_sorted":[],"title_canon_sha256":"21897dda10ec053a412daf05c6de0c8e398d193cd407f6d9ce5e2e8735ca979a","abstract_canon_sha256":"c6b5b36de25cb2aa8ff2681ce87bf2d33f2606acf350becd440c9fce4b85b941"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:15:36.694696Z","signature_b64":"01lXfjoKVrM5FQRQEyBczuUWXJcnXTP1qy2SQUmOTVQdl9zVa/Ni5wo9Ilrb6pZZqk94eMkJSXfZKWZGCGLjCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e07066ed0c098000dbfd294d30f87434bedd077da0250d6601fad2a8491dc8e7","last_reissued_at":"2026-07-05T00:15:36.694117Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:15:36.694117Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CARAFE: Content-Aware ReAssembly of FEatures","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Change Loy, Dahua Lin, Jiaqi Wang, Kai Chen, Rui Xu, Ziwei Liu","submitted_at":"2019-05-06T17:58:06Z","abstract_excerpt":"Feature upsampling is a key operation in a number of modern convolutional network architectures, e.g. feature pyramids. Its design is critical for dense prediction tasks such as object detection and semantic/instance segmentation. In this work, we propose Content-Aware ReAssembly of FEatures (CARAFE), a universal, lightweight and highly effective operator to fulfill this goal. CARAFE has several appealing properties: (1) Large field of view. Unlike previous works (e.g. bilinear interpolation) that only exploit sub-pixel neighborhood, CARAFE can aggregate contextual information within a large r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1905.02188","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/1905.02188/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":"1905.02188","created_at":"2026-07-05T00:15:36.694183+00:00"},{"alias_kind":"arxiv_version","alias_value":"1905.02188v3","created_at":"2026-07-05T00:15:36.694183+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1905.02188","created_at":"2026-07-05T00:15:36.694183+00:00"},{"alias_kind":"pith_short_12","alias_value":"4BYGN3IMBGAA","created_at":"2026-07-05T00:15:36.694183+00:00"},{"alias_kind":"pith_short_16","alias_value":"4BYGN3IMBGAABW75","created_at":"2026-07-05T00:15:36.694183+00:00"},{"alias_kind":"pith_short_8","alias_value":"4BYGN3IM","created_at":"2026-07-05T00:15:36.694183+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.06324","citing_title":"UniDemoir\\'e: Towards Universal Image Demoir\\'eing with Data Generation and Synthesis","ref_index":42,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS","json":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS.json","graph_json":"https://pith.science/api/pith-number/4BYGN3IMBGAABW75FFGTB6DUGS/graph.json","events_json":"https://pith.science/api/pith-number/4BYGN3IMBGAABW75FFGTB6DUGS/events.json","paper":"https://pith.science/paper/4BYGN3IM"},"agent_actions":{"view_html":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS","download_json":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS.json","view_paper":"https://pith.science/paper/4BYGN3IM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1905.02188&json=true","fetch_graph":"https://pith.science/api/pith-number/4BYGN3IMBGAABW75FFGTB6DUGS/graph.json","fetch_events":"https://pith.science/api/pith-number/4BYGN3IMBGAABW75FFGTB6DUGS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS/action/storage_attestation","attest_author":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS/action/author_attestation","sign_citation":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS/action/citation_signature","submit_replication":"https://pith.science/pith/4BYGN3IMBGAABW75FFGTB6DUGS/action/replication_record"}},"created_at":"2026-07-05T00:15:36.694183+00:00","updated_at":"2026-07-05T00:15:36.694183+00:00"}