{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:IYEOU42SSTLRJYWHJDBDUFKBJQ","short_pith_number":"pith:IYEOU42S","schema_version":"1.0","canonical_sha256":"4608ea735294d714e2c748c23a15414c1c37a2d4a8c79a185ca4095468e91543","source":{"kind":"arxiv","id":"2202.09844","version":3},"attestation_state":"computed","paper":{"title":"Sparsity Winning Twice: Better Robust Generalization from More Efficient Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Haoyu Ma, Pengjun Wang, Santosh Balachandra, Tianlong Chen, Zehao Wang, Zhangyang Wang, Zhenyu Zhang","submitted_at":"2022-02-20T15:52:08Z","abstract_excerpt":"Recent studies demonstrate that deep networks, even robustified by the state-of-the-art adversarial training (AT), still suffer from large robust generalization gaps, in addition to the much more expensive training costs than standard training. In this paper, we investigate this intriguing problem from a new perspective, i.e., injecting appropriate forms of sparsity during adversarial training. We introduce two alternatives for sparse adversarial training: (i) static sparsity, by leveraging recent results from the lottery ticket hypothesis to identify critical sparse subnetworks arising from t"},"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":"2202.09844","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-02-20T15:52:08Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0382fe88e777d5499794e8e7a72dc9fe70026f4a8cb3b922a4b0f222da3e9e93","abstract_canon_sha256":"9c74a1389ae181f8e1547da997d19a027c68c730681eeb4fb069367b77d49f18"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:00:22.991510Z","signature_b64":"Y2U6e5xP7XdGBwt15wwxW8sUrl5s7vMpjO+AHu5gVrr6Z/q72ke4ZZp/nQfnWmhgUxHqfV/BGQDMw5vM8wyBBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4608ea735294d714e2c748c23a15414c1c37a2d4a8c79a185ca4095468e91543","last_reissued_at":"2026-07-05T04:00:22.991157Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:00:22.991157Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sparsity Winning Twice: Better Robust Generalization from More Efficient Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Haoyu Ma, Pengjun Wang, Santosh Balachandra, Tianlong Chen, Zehao Wang, Zhangyang Wang, Zhenyu Zhang","submitted_at":"2022-02-20T15:52:08Z","abstract_excerpt":"Recent studies demonstrate that deep networks, even robustified by the state-of-the-art adversarial training (AT), still suffer from large robust generalization gaps, in addition to the much more expensive training costs than standard training. In this paper, we investigate this intriguing problem from a new perspective, i.e., injecting appropriate forms of sparsity during adversarial training. We introduce two alternatives for sparse adversarial training: (i) static sparsity, by leveraging recent results from the lottery ticket hypothesis to identify critical sparse subnetworks arising from t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.09844","kind":"arxiv","version":3},"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/2202.09844/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":"2202.09844","created_at":"2026-07-05T04:00:22.991213+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.09844v3","created_at":"2026-07-05T04:00:22.991213+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.09844","created_at":"2026-07-05T04:00:22.991213+00:00"},{"alias_kind":"pith_short_12","alias_value":"IYEOU42SSTLR","created_at":"2026-07-05T04:00:22.991213+00:00"},{"alias_kind":"pith_short_16","alias_value":"IYEOU42SSTLRJYWH","created_at":"2026-07-05T04:00:22.991213+00:00"},{"alias_kind":"pith_short_8","alias_value":"IYEOU42S","created_at":"2026-07-05T04:00:22.991213+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.11449","citing_title":"Dynamic Sparse Training of Diagonally Sparse Networks","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ","json":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ.json","graph_json":"https://pith.science/api/pith-number/IYEOU42SSTLRJYWHJDBDUFKBJQ/graph.json","events_json":"https://pith.science/api/pith-number/IYEOU42SSTLRJYWHJDBDUFKBJQ/events.json","paper":"https://pith.science/paper/IYEOU42S"},"agent_actions":{"view_html":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ","download_json":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ.json","view_paper":"https://pith.science/paper/IYEOU42S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.09844&json=true","fetch_graph":"https://pith.science/api/pith-number/IYEOU42SSTLRJYWHJDBDUFKBJQ/graph.json","fetch_events":"https://pith.science/api/pith-number/IYEOU42SSTLRJYWHJDBDUFKBJQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ/action/storage_attestation","attest_author":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ/action/author_attestation","sign_citation":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ/action/citation_signature","submit_replication":"https://pith.science/pith/IYEOU42SSTLRJYWHJDBDUFKBJQ/action/replication_record"}},"created_at":"2026-07-05T04:00:22.991213+00:00","updated_at":"2026-07-05T04:00:22.991213+00:00"}