{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:CZRPG77EJVFCJFHD6QXVVZ6SOP","short_pith_number":"pith:CZRPG77E","schema_version":"1.0","canonical_sha256":"1662f37fe44d4a2494e3f42f5ae7d273ea6f260bbaa3376b3e01ad60e377083e","source":{"kind":"arxiv","id":"2007.09654","version":4},"attestation_state":"computed","paper":{"title":"Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Qingqiu Huang, Tong Wu, Yu Wang, Ziwei Liu","submitted_at":"2020-07-19T11:50:10Z","abstract_excerpt":"We present a new loss function called Distribution-Balanced Loss for the multi-label recognition problems that exhibit long-tailed class distributions. Compared to conventional single-label classification problem, multi-label recognition problems are often more challenging due to two significant issues, namely the co-occurrence of labels and the dominance of negative labels (when treated as multiple binary classification problems). The Distribution-Balanced Loss tackles these issues through two key modifications to the standard binary cross-entropy loss: 1) a new way to re-balance the weights "},"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":"2007.09654","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-07-19T11:50:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5ef64864bad040880ff029fe9c45b9d45379dfbc7770fa5693f3b9d21f7e2963","abstract_canon_sha256":"f070b8edbf22c52a4f2ee24b6d24b632f5aae3f5085de023a62c50775f3c5690"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:37:38.240543Z","signature_b64":"lW6z7RM6XxXcosygktEn5dlyhGYAlTgsiMOpnHp0Ex6PTAIVP07Zu5q/LBsqCo9vrksCFAw9IWI+UjzUDezKAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1662f37fe44d4a2494e3f42f5ae7d273ea6f260bbaa3376b3e01ad60e377083e","last_reissued_at":"2026-07-05T03:37:38.240087Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:37:38.240087Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Qingqiu Huang, Tong Wu, Yu Wang, Ziwei Liu","submitted_at":"2020-07-19T11:50:10Z","abstract_excerpt":"We present a new loss function called Distribution-Balanced Loss for the multi-label recognition problems that exhibit long-tailed class distributions. Compared to conventional single-label classification problem, multi-label recognition problems are often more challenging due to two significant issues, namely the co-occurrence of labels and the dominance of negative labels (when treated as multiple binary classification problems). The Distribution-Balanced Loss tackles these issues through two key modifications to the standard binary cross-entropy loss: 1) a new way to re-balance the weights "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.09654","kind":"arxiv","version":4},"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/2007.09654/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":"2007.09654","created_at":"2026-07-05T03:37:38.240146+00:00"},{"alias_kind":"arxiv_version","alias_value":"2007.09654v4","created_at":"2026-07-05T03:37:38.240146+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.09654","created_at":"2026-07-05T03:37:38.240146+00:00"},{"alias_kind":"pith_short_12","alias_value":"CZRPG77EJVFC","created_at":"2026-07-05T03:37:38.240146+00:00"},{"alias_kind":"pith_short_16","alias_value":"CZRPG77EJVFCJFHD","created_at":"2026-07-05T03:37:38.240146+00:00"},{"alias_kind":"pith_short_8","alias_value":"CZRPG77E","created_at":"2026-07-05T03:37:38.240146+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.13059","citing_title":"Towards Unbiased and Robust Spatio-Temporal Scene Graph Generation and Anticipation","ref_index":50,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP","json":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP.json","graph_json":"https://pith.science/api/pith-number/CZRPG77EJVFCJFHD6QXVVZ6SOP/graph.json","events_json":"https://pith.science/api/pith-number/CZRPG77EJVFCJFHD6QXVVZ6SOP/events.json","paper":"https://pith.science/paper/CZRPG77E"},"agent_actions":{"view_html":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP","download_json":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP.json","view_paper":"https://pith.science/paper/CZRPG77E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2007.09654&json=true","fetch_graph":"https://pith.science/api/pith-number/CZRPG77EJVFCJFHD6QXVVZ6SOP/graph.json","fetch_events":"https://pith.science/api/pith-number/CZRPG77EJVFCJFHD6QXVVZ6SOP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP/action/storage_attestation","attest_author":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP/action/author_attestation","sign_citation":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP/action/citation_signature","submit_replication":"https://pith.science/pith/CZRPG77EJVFCJFHD6QXVVZ6SOP/action/replication_record"}},"created_at":"2026-07-05T03:37:38.240146+00:00","updated_at":"2026-07-05T03:37:38.240146+00:00"}