{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4DKYXQ6EAB2E5KLGOAP47E746J","short_pith_number":"pith:4DKYXQ6E","schema_version":"1.0","canonical_sha256":"e0d58bc3c400744ea966701fcf93fcf2750adc14385e11602b5d0f573d84f4f8","source":{"kind":"arxiv","id":"2310.13479","version":3},"attestation_state":"computed","paper":{"title":"Segment, Select, Correct: A Framework for Weakly-Supervised Referring Segmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Adel Bibi, Francisco Eiras, Kemal Oksuz, Philip H.S. Torr, Puneet K. Dokania","submitted_at":"2023-10-20T13:20:17Z","abstract_excerpt":"Referring Image Segmentation (RIS) - the problem of identifying objects in images through natural language sentences - is a challenging task currently mostly solved through supervised learning. However, while collecting referred annotation masks is a time-consuming process, the few existing weakly-supervised and zero-shot approaches fall significantly short in performance compared to fully-supervised learning ones. To bridge the performance gap without mask annotations, we propose a novel weakly-supervised framework that tackles RIS by decomposing it into three steps: obtaining instance masks "},"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":"2310.13479","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-10-20T13:20:17Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e514b0c53a654d9a45a386fb889da2ca4deb15da3a173cefef15e849f04230b2","abstract_canon_sha256":"fba34db28afbc65047b401f1a4f6728ff8ec0c96638a3579fff82d20f30ce09f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:56:57.729279Z","signature_b64":"JUYTXVowtLgaDhmSdFbyeC2zR2LedRw7dABPeAqhPVH/fBwz8rR/kMeHPF/7PQ1qc2ubv7kMSP1Y22sBowJ+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e0d58bc3c400744ea966701fcf93fcf2750adc14385e11602b5d0f573d84f4f8","last_reissued_at":"2026-07-05T08:56:57.728817Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:56:57.728817Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Segment, Select, Correct: A Framework for Weakly-Supervised Referring Segmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Adel Bibi, Francisco Eiras, Kemal Oksuz, Philip H.S. Torr, Puneet K. Dokania","submitted_at":"2023-10-20T13:20:17Z","abstract_excerpt":"Referring Image Segmentation (RIS) - the problem of identifying objects in images through natural language sentences - is a challenging task currently mostly solved through supervised learning. However, while collecting referred annotation masks is a time-consuming process, the few existing weakly-supervised and zero-shot approaches fall significantly short in performance compared to fully-supervised learning ones. To bridge the performance gap without mask annotations, we propose a novel weakly-supervised framework that tackles RIS by decomposing it into three steps: obtaining instance masks "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13479","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/2310.13479/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":"2310.13479","created_at":"2026-07-05T08:56:57.728867+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13479v3","created_at":"2026-07-05T08:56:57.728867+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13479","created_at":"2026-07-05T08:56:57.728867+00:00"},{"alias_kind":"pith_short_12","alias_value":"4DKYXQ6EAB2E","created_at":"2026-07-05T08:56:57.728867+00:00"},{"alias_kind":"pith_short_16","alias_value":"4DKYXQ6EAB2E5KLG","created_at":"2026-07-05T08:56:57.728867+00:00"},{"alias_kind":"pith_short_8","alias_value":"4DKYXQ6E","created_at":"2026-07-05T08:56:57.728867+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22546","citing_title":"Venice-H1: Failure-Aware Query Re-Ranking with Multi-Scale Grid Signatures for Referring Image Segmentation","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J","json":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J.json","graph_json":"https://pith.science/api/pith-number/4DKYXQ6EAB2E5KLGOAP47E746J/graph.json","events_json":"https://pith.science/api/pith-number/4DKYXQ6EAB2E5KLGOAP47E746J/events.json","paper":"https://pith.science/paper/4DKYXQ6E"},"agent_actions":{"view_html":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J","download_json":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J.json","view_paper":"https://pith.science/paper/4DKYXQ6E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13479&json=true","fetch_graph":"https://pith.science/api/pith-number/4DKYXQ6EAB2E5KLGOAP47E746J/graph.json","fetch_events":"https://pith.science/api/pith-number/4DKYXQ6EAB2E5KLGOAP47E746J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J/action/storage_attestation","attest_author":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J/action/author_attestation","sign_citation":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J/action/citation_signature","submit_replication":"https://pith.science/pith/4DKYXQ6EAB2E5KLGOAP47E746J/action/replication_record"}},"created_at":"2026-07-05T08:56:57.728867+00:00","updated_at":"2026-07-05T08:56:57.728867+00:00"}