{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:YHRJVKMWI37OFISOQTLXNLD7Z3","short_pith_number":"pith:YHRJVKMW","canonical_record":{"source":{"id":"2308.02533","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-08-01T09:02:34Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"73ea895cee7d5cf42dd2c9f04b59a9a121b3ab24511b0a693ad697f43b13c9b7","abstract_canon_sha256":"54739ab731e799643c64a4ee170bd2beb3966497a3d82eeeacdb2307f1183080"},"schema_version":"1.0"},"canonical_sha256":"c1e29aa99646fee2a24e84d776ac7fcef13742c49c230b7df769babdafe5881d","source":{"kind":"arxiv","id":"2308.02533","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.02533","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"arxiv_version","alias_value":"2308.02533v1","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02533","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"pith_short_12","alias_value":"YHRJVKMWI37O","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"pith_short_16","alias_value":"YHRJVKMWI37OFISO","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"pith_short_8","alias_value":"YHRJVKMW","created_at":"2026-07-05T06:38:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:YHRJVKMWI37OFISOQTLXNLD7Z3","target":"record","payload":{"canonical_record":{"source":{"id":"2308.02533","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-08-01T09:02:34Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"73ea895cee7d5cf42dd2c9f04b59a9a121b3ab24511b0a693ad697f43b13c9b7","abstract_canon_sha256":"54739ab731e799643c64a4ee170bd2beb3966497a3d82eeeacdb2307f1183080"},"schema_version":"1.0"},"canonical_sha256":"c1e29aa99646fee2a24e84d776ac7fcef13742c49c230b7df769babdafe5881d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:38:04.080676Z","signature_b64":"WILwG34TZzmGx9ZMOLqQwIAV/2TTt5W2Iu7kLvAw4YGCvuTJfQ1jAt6WtE9ljSjv9brYxcs2ub6HR3o6TezGDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c1e29aa99646fee2a24e84d776ac7fcef13742c49c230b7df769babdafe5881d","last_reissued_at":"2026-07-05T06:38:04.080335Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:38:04.080335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2308.02533","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:38:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AXxKLzq1ieIaObPengK6AbwVJYyaaFSyu4Ze8ybEBLoM9Liw8os6AaoQWBqAGHXgDKatBOKEb0TIk7vh2eflAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T09:16:16.261304Z"},"content_sha256":"2085cf344013ea8f6d345353be8e3ad96243ba5ab9bd8392dd8b105642ca7370","schema_version":"1.0","event_id":"sha256:2085cf344013ea8f6d345353be8e3ad96243ba5ab9bd8392dd8b105642ca7370"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:YHRJVKMWI37OFISOQTLXNLD7Z3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving Generalization of Adversarial Training via Robust Critical Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ge Yang, Jindong Wang, Kaijie Zhu, Xing Xie, Xixu Hu","submitted_at":"2023-08-01T09:02:34Z","abstract_excerpt":"Deep neural networks are susceptible to adversarial examples, posing a significant security risk in critical applications. Adversarial Training (AT) is a well-established technique to enhance adversarial robustness, but it often comes at the cost of decreased generalization ability. This paper proposes Robustness Critical Fine-Tuning (RiFT), a novel approach to enhance generalization without compromising adversarial robustness. The core idea of RiFT is to exploit the redundant capacity for robustness by fine-tuning the adversarially trained model on its non-robust-critical module. To do so, we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02533","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/2308.02533/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:38:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JndvAQGTtelNEI34XpNufpFF9S5nI8t33ac8oYT1V62ahpWsn/Sk0K81Tsg47+47bFyrX7XEtIjtCyR5Fz1ECQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T09:16:16.261703Z"},"content_sha256":"ae732647006a3296595fb6b2fedc1581af2676bd472d6324800de9136e0bedf8","schema_version":"1.0","event_id":"sha256:ae732647006a3296595fb6b2fedc1581af2676bd472d6324800de9136e0bedf8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YHRJVKMWI37OFISOQTLXNLD7Z3/bundle.json","state_url":"https://pith.science/pith/YHRJVKMWI37OFISOQTLXNLD7Z3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YHRJVKMWI37OFISOQTLXNLD7Z3/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-13T09:16:16Z","links":{"resolver":"https://pith.science/pith/YHRJVKMWI37OFISOQTLXNLD7Z3","bundle":"https://pith.science/pith/YHRJVKMWI37OFISOQTLXNLD7Z3/bundle.json","state":"https://pith.science/pith/YHRJVKMWI37OFISOQTLXNLD7Z3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YHRJVKMWI37OFISOQTLXNLD7Z3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YHRJVKMWI37OFISOQTLXNLD7Z3","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"54739ab731e799643c64a4ee170bd2beb3966497a3d82eeeacdb2307f1183080","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-08-01T09:02:34Z","title_canon_sha256":"73ea895cee7d5cf42dd2c9f04b59a9a121b3ab24511b0a693ad697f43b13c9b7"},"schema_version":"1.0","source":{"id":"2308.02533","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.02533","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"arxiv_version","alias_value":"2308.02533v1","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02533","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"pith_short_12","alias_value":"YHRJVKMWI37O","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"pith_short_16","alias_value":"YHRJVKMWI37OFISO","created_at":"2026-07-05T06:38:04Z"},{"alias_kind":"pith_short_8","alias_value":"YHRJVKMW","created_at":"2026-07-05T06:38:04Z"}],"graph_snapshots":[{"event_id":"sha256:ae732647006a3296595fb6b2fedc1581af2676bd472d6324800de9136e0bedf8","target":"graph","created_at":"2026-07-05T06:38:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2308.02533/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deep neural networks are susceptible to adversarial examples, posing a significant security risk in critical applications. Adversarial Training (AT) is a well-established technique to enhance adversarial robustness, but it often comes at the cost of decreased generalization ability. This paper proposes Robustness Critical Fine-Tuning (RiFT), a novel approach to enhance generalization without compromising adversarial robustness. The core idea of RiFT is to exploit the redundant capacity for robustness by fine-tuning the adversarially trained model on its non-robust-critical module. To do so, we","authors_text":"Ge Yang, Jindong Wang, Kaijie Zhu, Xing Xie, Xixu Hu","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-08-01T09:02:34Z","title":"Improving Generalization of Adversarial Training via Robust Critical Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02533","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2085cf344013ea8f6d345353be8e3ad96243ba5ab9bd8392dd8b105642ca7370","target":"record","created_at":"2026-07-05T06:38:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"54739ab731e799643c64a4ee170bd2beb3966497a3d82eeeacdb2307f1183080","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-08-01T09:02:34Z","title_canon_sha256":"73ea895cee7d5cf42dd2c9f04b59a9a121b3ab24511b0a693ad697f43b13c9b7"},"schema_version":"1.0","source":{"id":"2308.02533","kind":"arxiv","version":1}},"canonical_sha256":"c1e29aa99646fee2a24e84d776ac7fcef13742c49c230b7df769babdafe5881d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c1e29aa99646fee2a24e84d776ac7fcef13742c49c230b7df769babdafe5881d","first_computed_at":"2026-07-05T06:38:04.080335Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:38:04.080335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WILwG34TZzmGx9ZMOLqQwIAV/2TTt5W2Iu7kLvAw4YGCvuTJfQ1jAt6WtE9ljSjv9brYxcs2ub6HR3o6TezGDw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:38:04.080676Z","signed_message":"canonical_sha256_bytes"},"source_id":"2308.02533","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2085cf344013ea8f6d345353be8e3ad96243ba5ab9bd8392dd8b105642ca7370","sha256:ae732647006a3296595fb6b2fedc1581af2676bd472d6324800de9136e0bedf8"],"state_sha256":"6b7598b5edf5e88a9aa72e1d17171f816aabd5d0066a0ce881b5f8deb0ca5fec"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gUFzEiRK7NnweBCMAs8rawsEJO7aCSfwxSX9+T1QF9vBYXTVAp/0VrVVv0Bwdd34gL1Bng7Los7v8qZ0U+fnBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-13T09:16:16.263875Z","bundle_sha256":"475bd242d7cdf8e07f07cccfebc6aff2512b9257d6a0bec2dceddb1ce2d407fa"}}