{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:JT5RXT3UAIZHXR2ZKGK24UT63K","short_pith_number":"pith:JT5RXT3U","schema_version":"1.0","canonical_sha256":"4cfb1bcf7402327bc7595195ae527eda874960a23370fc2ddff98fb987fa4293","source":{"kind":"arxiv","id":"2206.01206","version":3},"attestation_state":"computed","paper":{"title":"Positive Unlabeled Contrastive Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anish Acharya, Bhargav Bhushanam, Dhruv Choudhary, Inderjit Dhillon, Li Jing, Michael Rabbat, Sujay Sanghavi","submitted_at":"2022-06-01T20:16:32Z","abstract_excerpt":"Self-supervised pretraining on unlabeled data followed by supervised fine-tuning on labeled data is a popular paradigm for learning from limited labeled examples. We extend this paradigm to the classical positive unlabeled (PU) setting, where the task is to learn a binary classifier given only a few labeled positive samples, and (often) a large amount of unlabeled samples (which could be positive or negative).\n  We first propose a simple extension of standard infoNCE family of contrastive losses, to the PU setting; and show that this learns superior representations, as compared to existing uns"},"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":"2206.01206","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-01T20:16:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3e02fe19172ca354ebfccd71ca2039aa66bc6c63dc4cee10429941725b214e00","abstract_canon_sha256":"f54288d13dda8cef4a619d1bcc24d8cfb8edd43594255501b8d904f9cd9b8fb1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:01:57.414114Z","signature_b64":"fP+wsjwoJzOzBu7R/iW16VcO/7nd1ZbTNPj7LV3O1XL8v0iu2/SkOfs+22FsUe3SwRQQtYy0ONKbaa4u5dIWAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cfb1bcf7402327bc7595195ae527eda874960a23370fc2ddff98fb987fa4293","last_reissued_at":"2026-07-05T08:01:57.413461Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:01:57.413461Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Positive Unlabeled Contrastive Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anish Acharya, Bhargav Bhushanam, Dhruv Choudhary, Inderjit Dhillon, Li Jing, Michael Rabbat, Sujay Sanghavi","submitted_at":"2022-06-01T20:16:32Z","abstract_excerpt":"Self-supervised pretraining on unlabeled data followed by supervised fine-tuning on labeled data is a popular paradigm for learning from limited labeled examples. We extend this paradigm to the classical positive unlabeled (PU) setting, where the task is to learn a binary classifier given only a few labeled positive samples, and (often) a large amount of unlabeled samples (which could be positive or negative).\n  We first propose a simple extension of standard infoNCE family of contrastive losses, to the PU setting; and show that this learns superior representations, as compared to existing uns"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.01206","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/2206.01206/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":"2206.01206","created_at":"2026-07-05T08:01:57.413543+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.01206v3","created_at":"2026-07-05T08:01:57.413543+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.01206","created_at":"2026-07-05T08:01:57.413543+00:00"},{"alias_kind":"pith_short_12","alias_value":"JT5RXT3UAIZH","created_at":"2026-07-05T08:01:57.413543+00:00"},{"alias_kind":"pith_short_16","alias_value":"JT5RXT3UAIZHXR2Z","created_at":"2026-07-05T08:01:57.413543+00:00"},{"alias_kind":"pith_short_8","alias_value":"JT5RXT3U","created_at":"2026-07-05T08:01:57.413543+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.06282","citing_title":"InfoNCE is a Free Lunch for Semantically guided Graph Contrastive Learning","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K","json":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K.json","graph_json":"https://pith.science/api/pith-number/JT5RXT3UAIZHXR2ZKGK24UT63K/graph.json","events_json":"https://pith.science/api/pith-number/JT5RXT3UAIZHXR2ZKGK24UT63K/events.json","paper":"https://pith.science/paper/JT5RXT3U"},"agent_actions":{"view_html":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K","download_json":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K.json","view_paper":"https://pith.science/paper/JT5RXT3U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.01206&json=true","fetch_graph":"https://pith.science/api/pith-number/JT5RXT3UAIZHXR2ZKGK24UT63K/graph.json","fetch_events":"https://pith.science/api/pith-number/JT5RXT3UAIZHXR2ZKGK24UT63K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K/action/storage_attestation","attest_author":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K/action/author_attestation","sign_citation":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K/action/citation_signature","submit_replication":"https://pith.science/pith/JT5RXT3UAIZHXR2ZKGK24UT63K/action/replication_record"}},"created_at":"2026-07-05T08:01:57.413543+00:00","updated_at":"2026-07-05T08:01:57.413543+00:00"}