{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QUBVLCWE5L2F2KIZLSD363L4GV","short_pith_number":"pith:QUBVLCWE","schema_version":"1.0","canonical_sha256":"8503558ac4eaf45d29195c87bf6d7c35542bee46455b5c721273ebaa0b9dbb73","source":{"kind":"arxiv","id":"2406.04070","version":1},"attestation_state":"computed","paper":{"title":"Batch-in-Batch: a new adversarial training framework for initial perturbation and sample selection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"(2) School of Mathematics, (3) Key Laboratory of Aerospace Information Security, Bo Cai (3), Central China Normal University, Computer Science, Engineering, Jianghan University, Key Lab NAA--MOE, Le Li (1). ((1) School of Mathematics, Ministry of Education, Pai Peng (2), School of Cyber Science, Statistics, Trusted Computing, Wuhan University), Yinting Wu (1)","submitted_at":"2024-06-06T13:34:43Z","abstract_excerpt":"Adversarial training methods commonly generate independent initial perturbation for adversarial samples from a simple uniform distribution, and obtain the training batch for the classifier without selection. In this work, we propose a simple yet effective training framework called Batch-in-Batch (BB) to enhance models robustness. It involves specifically a joint construction of initial values that could simultaneously generates $m$ sets of perturbations from the original batch set to provide more diversity for adversarial samples; and also includes various sample selection strategies that enab"},"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":"2406.04070","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-06T13:34:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e02daf7201c50a879005db3c27817618b37b7249b82d6abd2fe2ce803c961b48","abstract_canon_sha256":"414b634d5318e45798672a7df8ba9e4a13dd33cb3e4bd4ea4e0b285b4f0a2015"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:20.382964Z","signature_b64":"roiHo3J3JTYy/2XN51u4ll5Fu7jx+NsaJevuqNe3VZpWSqnzbmKPWBXtJtKOlO9PbGHZd3iewgmusDtsMNIrBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8503558ac4eaf45d29195c87bf6d7c35542bee46455b5c721273ebaa0b9dbb73","last_reissued_at":"2026-07-05T08:28:20.382513Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:20.382513Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Batch-in-Batch: a new adversarial training framework for initial perturbation and sample selection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"(2) School of Mathematics, (3) Key Laboratory of Aerospace Information Security, Bo Cai (3), Central China Normal University, Computer Science, Engineering, Jianghan University, Key Lab NAA--MOE, Le Li (1). ((1) School of Mathematics, Ministry of Education, Pai Peng (2), School of Cyber Science, Statistics, Trusted Computing, Wuhan University), Yinting Wu (1)","submitted_at":"2024-06-06T13:34:43Z","abstract_excerpt":"Adversarial training methods commonly generate independent initial perturbation for adversarial samples from a simple uniform distribution, and obtain the training batch for the classifier without selection. In this work, we propose a simple yet effective training framework called Batch-in-Batch (BB) to enhance models robustness. It involves specifically a joint construction of initial values that could simultaneously generates $m$ sets of perturbations from the original batch set to provide more diversity for adversarial samples; and also includes various sample selection strategies that enab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04070","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/2406.04070/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":"2406.04070","created_at":"2026-07-05T08:28:20.382570+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.04070v1","created_at":"2026-07-05T08:28:20.382570+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04070","created_at":"2026-07-05T08:28:20.382570+00:00"},{"alias_kind":"pith_short_12","alias_value":"QUBVLCWE5L2F","created_at":"2026-07-05T08:28:20.382570+00:00"},{"alias_kind":"pith_short_16","alias_value":"QUBVLCWE5L2F2KIZ","created_at":"2026-07-05T08:28:20.382570+00:00"},{"alias_kind":"pith_short_8","alias_value":"QUBVLCWE","created_at":"2026-07-05T08:28:20.382570+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/QUBVLCWE5L2F2KIZLSD363L4GV","json":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV.json","graph_json":"https://pith.science/api/pith-number/QUBVLCWE5L2F2KIZLSD363L4GV/graph.json","events_json":"https://pith.science/api/pith-number/QUBVLCWE5L2F2KIZLSD363L4GV/events.json","paper":"https://pith.science/paper/QUBVLCWE"},"agent_actions":{"view_html":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV","download_json":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV.json","view_paper":"https://pith.science/paper/QUBVLCWE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.04070&json=true","fetch_graph":"https://pith.science/api/pith-number/QUBVLCWE5L2F2KIZLSD363L4GV/graph.json","fetch_events":"https://pith.science/api/pith-number/QUBVLCWE5L2F2KIZLSD363L4GV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV/action/storage_attestation","attest_author":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV/action/author_attestation","sign_citation":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV/action/citation_signature","submit_replication":"https://pith.science/pith/QUBVLCWE5L2F2KIZLSD363L4GV/action/replication_record"}},"created_at":"2026-07-05T08:28:20.382570+00:00","updated_at":"2026-07-05T08:28:20.382570+00:00"}