{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LNNDYNSZ5PC5WOJZJWMQ6DREWT","short_pith_number":"pith:LNNDYNSZ","schema_version":"1.0","canonical_sha256":"5b5a3c3659ebc5db39394d990f0e24b4f3624bb41caaad2a91136369d344c4cc","source":{"kind":"arxiv","id":"2508.03579","version":2},"attestation_state":"computed","paper":{"title":"Heterogeneity-Oblivious Robust Federated Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NI"],"primary_cat":"cs.LG","authors_text":"Chungang Lin, Haitong Luo, Jinyang Li, Miao Wang, Qi Song, Weiyao Zhang, Xuying Meng, Yujun Zhang","submitted_at":"2025-08-05T15:47:16Z","abstract_excerpt":"Federated Learning (FL) remains highly vulnerable to poisoning attacks, especially under real-world hyper-heterogeneity, where clients differ significantly in data distributions, communication capabilities, and model architectures. Such heterogeneity not only undermines the effectiveness of aggregation strategies but also makes attacks more difficult to detect. Furthermore, high-dimensional models expand the attack surface. To address these challenges, we propose Horus, a heterogeneity-oblivious robust FL framework centered on low-rank adaptations (LoRAs). Rather than aggregating full model pa"},"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":"2508.03579","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-05T15:47:16Z","cross_cats_sorted":["cs.NI"],"title_canon_sha256":"8bce7924cc91770f0fcf72ec5564822f88f9a98f6fb1fd63fff699757b5f0d58","abstract_canon_sha256":"059a0af1e2c3310d5d360bf326cb5dc1252cb7b3f88fad6b54ce431209ead547"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:49:13.084968Z","signature_b64":"+rahmxHFrViZ9USadAHsCSHuCBQfer0bcH9t5j7vPGeb8UVjJzkyHddwTYVyX4ZtV8hyC4F71KRSHuc2riUgCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5b5a3c3659ebc5db39394d990f0e24b4f3624bb41caaad2a91136369d344c4cc","last_reissued_at":"2026-07-05T11:49:13.084509Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:49:13.084509Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Heterogeneity-Oblivious Robust Federated Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NI"],"primary_cat":"cs.LG","authors_text":"Chungang Lin, Haitong Luo, Jinyang Li, Miao Wang, Qi Song, Weiyao Zhang, Xuying Meng, Yujun Zhang","submitted_at":"2025-08-05T15:47:16Z","abstract_excerpt":"Federated Learning (FL) remains highly vulnerable to poisoning attacks, especially under real-world hyper-heterogeneity, where clients differ significantly in data distributions, communication capabilities, and model architectures. Such heterogeneity not only undermines the effectiveness of aggregation strategies but also makes attacks more difficult to detect. Furthermore, high-dimensional models expand the attack surface. To address these challenges, we propose Horus, a heterogeneity-oblivious robust FL framework centered on low-rank adaptations (LoRAs). Rather than aggregating full model pa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.03579","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/2508.03579/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":"2508.03579","created_at":"2026-07-05T11:49:13.084572+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.03579v2","created_at":"2026-07-05T11:49:13.084572+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.03579","created_at":"2026-07-05T11:49:13.084572+00:00"},{"alias_kind":"pith_short_12","alias_value":"LNNDYNSZ5PC5","created_at":"2026-07-05T11:49:13.084572+00:00"},{"alias_kind":"pith_short_16","alias_value":"LNNDYNSZ5PC5WOJZ","created_at":"2026-07-05T11:49:13.084572+00:00"},{"alias_kind":"pith_short_8","alias_value":"LNNDYNSZ","created_at":"2026-07-05T11:49:13.084572+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/LNNDYNSZ5PC5WOJZJWMQ6DREWT","json":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT.json","graph_json":"https://pith.science/api/pith-number/LNNDYNSZ5PC5WOJZJWMQ6DREWT/graph.json","events_json":"https://pith.science/api/pith-number/LNNDYNSZ5PC5WOJZJWMQ6DREWT/events.json","paper":"https://pith.science/paper/LNNDYNSZ"},"agent_actions":{"view_html":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT","download_json":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT.json","view_paper":"https://pith.science/paper/LNNDYNSZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.03579&json=true","fetch_graph":"https://pith.science/api/pith-number/LNNDYNSZ5PC5WOJZJWMQ6DREWT/graph.json","fetch_events":"https://pith.science/api/pith-number/LNNDYNSZ5PC5WOJZJWMQ6DREWT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT/action/storage_attestation","attest_author":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT/action/author_attestation","sign_citation":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT/action/citation_signature","submit_replication":"https://pith.science/pith/LNNDYNSZ5PC5WOJZJWMQ6DREWT/action/replication_record"}},"created_at":"2026-07-05T11:49:13.084572+00:00","updated_at":"2026-07-05T11:49:13.084572+00:00"}