{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:GO3DDWLK7IALJWB2ZYYNHEYK4D","short_pith_number":"pith:GO3DDWLK","schema_version":"1.0","canonical_sha256":"33b631d96afa00b4d83ace30d3930ae0d4a008bef2ad64dff35cd241074a223b","source":{"kind":"arxiv","id":"2205.14969","version":3},"attestation_state":"computed","paper":{"title":"Guided Diffusion Model for Adversarial Purification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bo Dai, Dahua Lin, Hongfei Fu, Jinyi Wang, Zhaoyang Lyu","submitted_at":"2022-05-30T10:11:15Z","abstract_excerpt":"With wider application of deep neural networks (DNNs) in various algorithms and frameworks, security threats have become one of the concerns. Adversarial attacks disturb DNN-based image classifiers, in which attackers can intentionally add imperceptible adversarial perturbations on input images to fool the classifiers. In this paper, we propose a novel purification approach, referred to as guided diffusion model for purification (GDMP), to help protect classifiers from adversarial attacks. The core of our approach is to embed purification into the diffusion denoising process of a Denoised Diff"},"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":"2205.14969","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-30T10:11:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4b63063a726bf1adb7c2b6cc21bd0f36234eab227427b112d022580a281b955d","abstract_canon_sha256":"dcba4f39cbbb899bf009031c9b9ec917cb378d58ef76b5980b2884003f6eb1a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:36:00.891627Z","signature_b64":"DUabtKEIBMLjmC3udM9GU1UoJTgAaBXf+hc3HDXJmX2E2ktB8sUWEXPhsAq4pUEdJcHHFkX5wCMCcbMej9nlCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"33b631d96afa00b4d83ace30d3930ae0d4a008bef2ad64dff35cd241074a223b","last_reissued_at":"2026-07-05T04:36:00.891142Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:36:00.891142Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Guided Diffusion Model for Adversarial Purification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bo Dai, Dahua Lin, Hongfei Fu, Jinyi Wang, Zhaoyang Lyu","submitted_at":"2022-05-30T10:11:15Z","abstract_excerpt":"With wider application of deep neural networks (DNNs) in various algorithms and frameworks, security threats have become one of the concerns. Adversarial attacks disturb DNN-based image classifiers, in which attackers can intentionally add imperceptible adversarial perturbations on input images to fool the classifiers. In this paper, we propose a novel purification approach, referred to as guided diffusion model for purification (GDMP), to help protect classifiers from adversarial attacks. The core of our approach is to embed purification into the diffusion denoising process of a Denoised Diff"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.14969","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/2205.14969/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":"2205.14969","created_at":"2026-07-05T04:36:00.891199+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.14969v3","created_at":"2026-07-05T04:36:00.891199+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.14969","created_at":"2026-07-05T04:36:00.891199+00:00"},{"alias_kind":"pith_short_12","alias_value":"GO3DDWLK7IAL","created_at":"2026-07-05T04:36:00.891199+00:00"},{"alias_kind":"pith_short_16","alias_value":"GO3DDWLK7IALJWB2","created_at":"2026-07-05T04:36:00.891199+00:00"},{"alias_kind":"pith_short_8","alias_value":"GO3DDWLK","created_at":"2026-07-05T04:36:00.891199+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01305","citing_title":"Generative AI and Federated Learning for Intrusion Detection Systems: A Survey","ref_index":97,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08745","citing_title":"Stain-Aware Wavelet Regularization for Instant Adversarial Purification in Histopathology","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03730","citing_title":"Beyond False Stability: High-Noise Drift Gating for Test-Time Adversarial Defenses in Vision-Language Models","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08440","citing_title":"TARO: Temporal Adversarial Rectification Optimization Using Diffusion Models as Purifiers","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D","json":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D.json","graph_json":"https://pith.science/api/pith-number/GO3DDWLK7IALJWB2ZYYNHEYK4D/graph.json","events_json":"https://pith.science/api/pith-number/GO3DDWLK7IALJWB2ZYYNHEYK4D/events.json","paper":"https://pith.science/paper/GO3DDWLK"},"agent_actions":{"view_html":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D","download_json":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D.json","view_paper":"https://pith.science/paper/GO3DDWLK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.14969&json=true","fetch_graph":"https://pith.science/api/pith-number/GO3DDWLK7IALJWB2ZYYNHEYK4D/graph.json","fetch_events":"https://pith.science/api/pith-number/GO3DDWLK7IALJWB2ZYYNHEYK4D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D/action/storage_attestation","attest_author":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D/action/author_attestation","sign_citation":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D/action/citation_signature","submit_replication":"https://pith.science/pith/GO3DDWLK7IALJWB2ZYYNHEYK4D/action/replication_record"}},"created_at":"2026-07-05T04:36:00.891199+00:00","updated_at":"2026-07-05T04:36:00.891199+00:00"}