{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:5OFKTDZ7D5HCKYJELFNSWBHWJN","short_pith_number":"pith:5OFKTDZ7","schema_version":"1.0","canonical_sha256":"eb8aa98f3f1f4e256124595b2b04f64b63f5dcf5a3ef82d27c2578d9fc03e289","source":{"kind":"arxiv","id":"2303.15014","version":1},"attestation_state":"computed","paper":{"title":"Leveraging Hidden Positives for Unsupervised Semantic Segmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hyun Seok Seong, Jae-Pil Heo, SuBeen Lee, WonJun Moon","submitted_at":"2023-03-27T08:57:28Z","abstract_excerpt":"Dramatic demand for manpower to label pixel-level annotations triggered the advent of unsupervised semantic segmentation. Although the recent work employing the vision transformer (ViT) backbone shows exceptional performance, there is still a lack of consideration for task-specific training guidance and local semantic consistency. To tackle these issues, we leverage contrastive learning by excavating hidden positives to learn rich semantic relationships and ensure semantic consistency in local regions. Specifically, we first discover two types of global hidden positives, task-agnostic and task"},"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":"2303.15014","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-27T08:57:28Z","cross_cats_sorted":[],"title_canon_sha256":"5f879de7a4b018f627c2f61fe0b39887ba84daf3d0f32d5e4cbcea7f54961589","abstract_canon_sha256":"ea91400b23920a44dfbc252d1fe1bb5801b0d05f1284d26ae4df92c66f10cdb9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:55:00.067353Z","signature_b64":"BDVLT4xxUEIORi7Ryd4lItbKgijpxQ2+KOVxHLiMjJo9cfInPg89K6KWUgmcieq76GKoibfR3C0g0XBxv5HsBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eb8aa98f3f1f4e256124595b2b04f64b63f5dcf5a3ef82d27c2578d9fc03e289","last_reissued_at":"2026-07-05T05:55:00.066897Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:55:00.066897Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Leveraging Hidden Positives for Unsupervised Semantic Segmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hyun Seok Seong, Jae-Pil Heo, SuBeen Lee, WonJun Moon","submitted_at":"2023-03-27T08:57:28Z","abstract_excerpt":"Dramatic demand for manpower to label pixel-level annotations triggered the advent of unsupervised semantic segmentation. Although the recent work employing the vision transformer (ViT) backbone shows exceptional performance, there is still a lack of consideration for task-specific training guidance and local semantic consistency. To tackle these issues, we leverage contrastive learning by excavating hidden positives to learn rich semantic relationships and ensure semantic consistency in local regions. Specifically, we first discover two types of global hidden positives, task-agnostic and task"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.15014","kind":"arxiv","version":1},"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/2303.15014/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":"2303.15014","created_at":"2026-07-05T05:55:00.066954+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.15014v1","created_at":"2026-07-05T05:55:00.066954+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.15014","created_at":"2026-07-05T05:55:00.066954+00:00"},{"alias_kind":"pith_short_12","alias_value":"5OFKTDZ7D5HC","created_at":"2026-07-05T05:55:00.066954+00:00"},{"alias_kind":"pith_short_16","alias_value":"5OFKTDZ7D5HCKYJE","created_at":"2026-07-05T05:55:00.066954+00:00"},{"alias_kind":"pith_short_8","alias_value":"5OFKTDZ7","created_at":"2026-07-05T05:55:00.066954+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/5OFKTDZ7D5HCKYJELFNSWBHWJN","json":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN.json","graph_json":"https://pith.science/api/pith-number/5OFKTDZ7D5HCKYJELFNSWBHWJN/graph.json","events_json":"https://pith.science/api/pith-number/5OFKTDZ7D5HCKYJELFNSWBHWJN/events.json","paper":"https://pith.science/paper/5OFKTDZ7"},"agent_actions":{"view_html":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN","download_json":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN.json","view_paper":"https://pith.science/paper/5OFKTDZ7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.15014&json=true","fetch_graph":"https://pith.science/api/pith-number/5OFKTDZ7D5HCKYJELFNSWBHWJN/graph.json","fetch_events":"https://pith.science/api/pith-number/5OFKTDZ7D5HCKYJELFNSWBHWJN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN/action/storage_attestation","attest_author":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN/action/author_attestation","sign_citation":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN/action/citation_signature","submit_replication":"https://pith.science/pith/5OFKTDZ7D5HCKYJELFNSWBHWJN/action/replication_record"}},"created_at":"2026-07-05T05:55:00.066954+00:00","updated_at":"2026-07-05T05:55:00.066954+00:00"}