{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:BV7M64LJ4HFQELQ2GFUL6Z4GNK","short_pith_number":"pith:BV7M64LJ","schema_version":"1.0","canonical_sha256":"0d7ecf7169e1cb022e1a3168bf67866a99300e74ef7281d402ed18dbf39b53cd","source":{"kind":"arxiv","id":"2003.13911","version":1},"attestation_state":"computed","paper":{"title":"Proxy Anchor Loss for Deep Metric Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dongwon Kim, Minsu Cho, Suha Kwak, Sungyeon Kim","submitted_at":"2020-03-31T02:05:27Z","abstract_excerpt":"Existing metric learning losses can be categorized into two classes: pair-based and proxy-based losses. The former class can leverage fine-grained semantic relations between data points, but slows convergence in general due to its high training complexity. In contrast, the latter class enables fast and reliable convergence, but cannot consider the rich data-to-data relations. This paper presents a new proxy-based loss that takes advantages of both pair- and proxy-based methods and overcomes their limitations. Thanks to the use of proxies, our loss boosts the speed of convergence and is robust "},"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":"2003.13911","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-03-31T02:05:27Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"eecd1732e79797307888529975d843dbf7c2143f99d73a6a25896b63befab503","abstract_canon_sha256":"b1a13be31a3ac375189dfc8af6133faa54cacc14f985357156cd9c2c4a050fc1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:51:46.294053Z","signature_b64":"fqLmYYxvCNlC4ILYU9GwcAmXSE0OId4wdmKPuO2HP3VRmXErpGtmDpiYbkzqnJE70Jar0M/nIDELLKqJL7qOAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d7ecf7169e1cb022e1a3168bf67866a99300e74ef7281d402ed18dbf39b53cd","last_reissued_at":"2026-07-05T00:51:46.293561Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:51:46.293561Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Proxy Anchor Loss for Deep Metric Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dongwon Kim, Minsu Cho, Suha Kwak, Sungyeon Kim","submitted_at":"2020-03-31T02:05:27Z","abstract_excerpt":"Existing metric learning losses can be categorized into two classes: pair-based and proxy-based losses. The former class can leverage fine-grained semantic relations between data points, but slows convergence in general due to its high training complexity. In contrast, the latter class enables fast and reliable convergence, but cannot consider the rich data-to-data relations. This paper presents a new proxy-based loss that takes advantages of both pair- and proxy-based methods and overcomes their limitations. Thanks to the use of proxies, our loss boosts the speed of convergence and is robust "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.13911","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/2003.13911/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":"2003.13911","created_at":"2026-07-05T00:51:46.293619+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.13911v1","created_at":"2026-07-05T00:51:46.293619+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.13911","created_at":"2026-07-05T00:51:46.293619+00:00"},{"alias_kind":"pith_short_12","alias_value":"BV7M64LJ4HFQ","created_at":"2026-07-05T00:51:46.293619+00:00"},{"alias_kind":"pith_short_16","alias_value":"BV7M64LJ4HFQELQ2","created_at":"2026-07-05T00:51:46.293619+00:00"},{"alias_kind":"pith_short_8","alias_value":"BV7M64LJ","created_at":"2026-07-05T00:51:46.293619+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10758","citing_title":"Anchoring the Unknown: Open-Set Model Attribution via Proxy-Anchor Learning","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11654","citing_title":"Weather-Robust Cross-View Geo-Localization via Prototype-Based Semantic Part Discovery","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11654","citing_title":"Weather-Robust Cross-View Geo-Localization via Prototype-Based Semantic Part Discovery","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK","json":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK.json","graph_json":"https://pith.science/api/pith-number/BV7M64LJ4HFQELQ2GFUL6Z4GNK/graph.json","events_json":"https://pith.science/api/pith-number/BV7M64LJ4HFQELQ2GFUL6Z4GNK/events.json","paper":"https://pith.science/paper/BV7M64LJ"},"agent_actions":{"view_html":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK","download_json":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK.json","view_paper":"https://pith.science/paper/BV7M64LJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.13911&json=true","fetch_graph":"https://pith.science/api/pith-number/BV7M64LJ4HFQELQ2GFUL6Z4GNK/graph.json","fetch_events":"https://pith.science/api/pith-number/BV7M64LJ4HFQELQ2GFUL6Z4GNK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK/action/storage_attestation","attest_author":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK/action/author_attestation","sign_citation":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK/action/citation_signature","submit_replication":"https://pith.science/pith/BV7M64LJ4HFQELQ2GFUL6Z4GNK/action/replication_record"}},"created_at":"2026-07-05T00:51:46.293619+00:00","updated_at":"2026-07-05T00:51:46.293619+00:00"}