{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WTWD6IPD3SX7AW6OULXVEBAKAV","short_pith_number":"pith:WTWD6IPD","schema_version":"1.0","canonical_sha256":"b4ec3f21e3dcaff05bcea2ef52040a0557ec0329f829807b4bc0b887c9b161df","source":{"kind":"arxiv","id":"2312.12703","version":2},"attestation_state":"computed","paper":{"title":"Federated Learning with Extremely Noisy Clients via Negative Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Han, Hanzi Wang, Lin Chen, Yang Lu, Yiliang Zhang, Yiu-ming Cheung, Yonggang Zhang","submitted_at":"2023-12-20T01:59:48Z","abstract_excerpt":"Federated learning (FL) has shown remarkable success in cooperatively training deep models, while typically struggling with noisy labels. Advanced works propose to tackle label noise by a re-weighting strategy with a strong assumption, i.e., mild label noise. However, it may be violated in many real-world FL scenarios because of highly contaminated clients, resulting in extreme noise ratios, e.g., $>$90%. To tackle extremely noisy clients, we study the robustness of the re-weighting strategy, showing a pessimistic conclusion: minimizing the weight of clients trained over noisy data outperforms"},"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":"2312.12703","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-12-20T01:59:48Z","cross_cats_sorted":[],"title_canon_sha256":"c0301d192a0b89a3759e826f3478c5f31d98e761f408c6d83d5a18d0f40d42ae","abstract_canon_sha256":"b09bf602d6524e541b74b9d37102e5713b792ba9f668b59ecd1801e75e803a87"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:48:33.405897Z","signature_b64":"zoWyFSibnRQN09H9Sreo4vhQCiulEP4sLfLn2wYIdCF974o/RXI7jiGFHvnwkDFIe6J8TPQAk1vCRGm0KC2ICQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4ec3f21e3dcaff05bcea2ef52040a0557ec0329f829807b4bc0b887c9b161df","last_reissued_at":"2026-07-05T07:48:33.405448Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:48:33.405448Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Federated Learning with Extremely Noisy Clients via Negative Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Han, Hanzi Wang, Lin Chen, Yang Lu, Yiliang Zhang, Yiu-ming Cheung, Yonggang Zhang","submitted_at":"2023-12-20T01:59:48Z","abstract_excerpt":"Federated learning (FL) has shown remarkable success in cooperatively training deep models, while typically struggling with noisy labels. Advanced works propose to tackle label noise by a re-weighting strategy with a strong assumption, i.e., mild label noise. However, it may be violated in many real-world FL scenarios because of highly contaminated clients, resulting in extreme noise ratios, e.g., $>$90%. To tackle extremely noisy clients, we study the robustness of the re-weighting strategy, showing a pessimistic conclusion: minimizing the weight of clients trained over noisy data outperforms"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.12703","kind":"arxiv","version":2},"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/2312.12703/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":"2312.12703","created_at":"2026-07-05T07:48:33.405504+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.12703v2","created_at":"2026-07-05T07:48:33.405504+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.12703","created_at":"2026-07-05T07:48:33.405504+00:00"},{"alias_kind":"pith_short_12","alias_value":"WTWD6IPD3SX7","created_at":"2026-07-05T07:48:33.405504+00:00"},{"alias_kind":"pith_short_16","alias_value":"WTWD6IPD3SX7AW6O","created_at":"2026-07-05T07:48:33.405504+00:00"},{"alias_kind":"pith_short_8","alias_value":"WTWD6IPD","created_at":"2026-07-05T07:48:33.405504+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.02079","citing_title":"Robust Federated Learning against Noisy Clients via Masked Optimization","ref_index":35,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV","json":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV.json","graph_json":"https://pith.science/api/pith-number/WTWD6IPD3SX7AW6OULXVEBAKAV/graph.json","events_json":"https://pith.science/api/pith-number/WTWD6IPD3SX7AW6OULXVEBAKAV/events.json","paper":"https://pith.science/paper/WTWD6IPD"},"agent_actions":{"view_html":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV","download_json":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV.json","view_paper":"https://pith.science/paper/WTWD6IPD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.12703&json=true","fetch_graph":"https://pith.science/api/pith-number/WTWD6IPD3SX7AW6OULXVEBAKAV/graph.json","fetch_events":"https://pith.science/api/pith-number/WTWD6IPD3SX7AW6OULXVEBAKAV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV/action/storage_attestation","attest_author":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV/action/author_attestation","sign_citation":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV/action/citation_signature","submit_replication":"https://pith.science/pith/WTWD6IPD3SX7AW6OULXVEBAKAV/action/replication_record"}},"created_at":"2026-07-05T07:48:33.405504+00:00","updated_at":"2026-07-05T07:48:33.405504+00:00"}