{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:3IANW3PFTNN2STBZEXLWHJPQYZ","short_pith_number":"pith:3IANW3PF","canonical_record":{"source":{"id":"2110.07801","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-15T01:45:31Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"89ba6c93f0fa78e067ec15a0a3b9823f959c71059cc8279ae6a7588fc897d5d8","abstract_canon_sha256":"9f51dd902be47457531cc0e09db6d0b87ab7253e7dc753ef629e4fb36a103783"},"schema_version":"1.0"},"canonical_sha256":"da00db6de59b5ba94c3925d763a5f0c66478303b71abbdb846502288b521dbde","source":{"kind":"arxiv","id":"2110.07801","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.07801","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"arxiv_version","alias_value":"2110.07801v1","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.07801","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"pith_short_12","alias_value":"3IANW3PFTNN2","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"pith_short_16","alias_value":"3IANW3PFTNN2STBZ","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"pith_short_8","alias_value":"3IANW3PF","created_at":"2026-07-05T03:22:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:3IANW3PFTNN2STBZEXLWHJPQYZ","target":"record","payload":{"canonical_record":{"source":{"id":"2110.07801","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-15T01:45:31Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"89ba6c93f0fa78e067ec15a0a3b9823f959c71059cc8279ae6a7588fc897d5d8","abstract_canon_sha256":"9f51dd902be47457531cc0e09db6d0b87ab7253e7dc753ef629e4fb36a103783"},"schema_version":"1.0"},"canonical_sha256":"da00db6de59b5ba94c3925d763a5f0c66478303b71abbdb846502288b521dbde","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:22:49.952457Z","signature_b64":"fmM9ryUKe/z5VIiDXBsH0xoeo0R2jk/N+pTcQuo/ahvGxYm1ayyYTiz4lo6RW4FA8NuOuJ2bIrhnGJ1+35MyBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da00db6de59b5ba94c3925d763a5f0c66478303b71abbdb846502288b521dbde","last_reissued_at":"2026-07-05T03:22:49.952030Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:22:49.952030Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2110.07801","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":"2026-07-05T03:22:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N+dOCjiGAgIDDd6/yJ3laoWwjPLlS3tTH/EKIkuaBe2oNEJPM+BQ6Nsr1YNU2MMniSdFSBohrjJcussXQlOxBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:17:36.748654Z"},"content_sha256":"c6408fe47102e64a5977b67cc65552fa3c7db2da12fab06ba50b7da8476bae29","schema_version":"1.0","event_id":"sha256:c6408fe47102e64a5977b67cc65552fa3c7db2da12fab06ba50b7da8476bae29"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:3IANW3PFTNN2STBZEXLWHJPQYZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adversarial Purification through Representation Disentanglement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Bihan Wen, Jun Zhao, Lanqing Guo, Tao Bai","submitted_at":"2021-10-15T01:45:31Z","abstract_excerpt":"Deep learning models are vulnerable to adversarial examples and make incomprehensible mistakes, which puts a threat on their real-world deployment. Combined with the idea of adversarial training, preprocessing-based defenses are popular and convenient to use because of their task independence and good generalizability. Current defense methods, especially purification, tend to remove ``noise\" by learning and recovering the natural images. However, different from random noise, the adversarial patterns are much easier to be overfitted during model training due to their strong correlation to the i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.07801","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/2110.07801/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":"2026-07-05T03:22:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IY9Qd6a/ALHzcyCrLRdpxvgVwz51EyXNHDGX8c2G7n2qn6aYinTemuiQQye35Jbwrj8Rlo/aAokVN9IM71orBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T09:17:36.749027Z"},"content_sha256":"a40a0b98d6a75ec3dc881a07c99b31852fe9561a6c0f2b8cb30fa8b2fdbeac81","schema_version":"1.0","event_id":"sha256:a40a0b98d6a75ec3dc881a07c99b31852fe9561a6c0f2b8cb30fa8b2fdbeac81"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3IANW3PFTNN2STBZEXLWHJPQYZ/bundle.json","state_url":"https://pith.science/pith/3IANW3PFTNN2STBZEXLWHJPQYZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3IANW3PFTNN2STBZEXLWHJPQYZ/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":"2026-07-07T09:17:36Z","links":{"resolver":"https://pith.science/pith/3IANW3PFTNN2STBZEXLWHJPQYZ","bundle":"https://pith.science/pith/3IANW3PFTNN2STBZEXLWHJPQYZ/bundle.json","state":"https://pith.science/pith/3IANW3PFTNN2STBZEXLWHJPQYZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3IANW3PFTNN2STBZEXLWHJPQYZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:3IANW3PFTNN2STBZEXLWHJPQYZ","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":"9f51dd902be47457531cc0e09db6d0b87ab7253e7dc753ef629e4fb36a103783","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-15T01:45:31Z","title_canon_sha256":"89ba6c93f0fa78e067ec15a0a3b9823f959c71059cc8279ae6a7588fc897d5d8"},"schema_version":"1.0","source":{"id":"2110.07801","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2110.07801","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"arxiv_version","alias_value":"2110.07801v1","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.07801","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"pith_short_12","alias_value":"3IANW3PFTNN2","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"pith_short_16","alias_value":"3IANW3PFTNN2STBZ","created_at":"2026-07-05T03:22:49Z"},{"alias_kind":"pith_short_8","alias_value":"3IANW3PF","created_at":"2026-07-05T03:22:49Z"}],"graph_snapshots":[{"event_id":"sha256:a40a0b98d6a75ec3dc881a07c99b31852fe9561a6c0f2b8cb30fa8b2fdbeac81","target":"graph","created_at":"2026-07-05T03:22:49Z","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/2110.07801/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Deep learning models are vulnerable to adversarial examples and make incomprehensible mistakes, which puts a threat on their real-world deployment. Combined with the idea of adversarial training, preprocessing-based defenses are popular and convenient to use because of their task independence and good generalizability. Current defense methods, especially purification, tend to remove ``noise\" by learning and recovering the natural images. However, different from random noise, the adversarial patterns are much easier to be overfitted during model training due to their strong correlation to the i","authors_text":"Bihan Wen, Jun Zhao, Lanqing Guo, Tao Bai","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-15T01:45:31Z","title":"Adversarial Purification through Representation Disentanglement"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.07801","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:c6408fe47102e64a5977b67cc65552fa3c7db2da12fab06ba50b7da8476bae29","target":"record","created_at":"2026-07-05T03:22:49Z","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":"9f51dd902be47457531cc0e09db6d0b87ab7253e7dc753ef629e4fb36a103783","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-15T01:45:31Z","title_canon_sha256":"89ba6c93f0fa78e067ec15a0a3b9823f959c71059cc8279ae6a7588fc897d5d8"},"schema_version":"1.0","source":{"id":"2110.07801","kind":"arxiv","version":1}},"canonical_sha256":"da00db6de59b5ba94c3925d763a5f0c66478303b71abbdb846502288b521dbde","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da00db6de59b5ba94c3925d763a5f0c66478303b71abbdb846502288b521dbde","first_computed_at":"2026-07-05T03:22:49.952030Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:22:49.952030Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fmM9ryUKe/z5VIiDXBsH0xoeo0R2jk/N+pTcQuo/ahvGxYm1ayyYTiz4lo6RW4FA8NuOuJ2bIrhnGJ1+35MyBw==","signature_status":"signed_v1","signed_at":"2026-07-05T03:22:49.952457Z","signed_message":"canonical_sha256_bytes"},"source_id":"2110.07801","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c6408fe47102e64a5977b67cc65552fa3c7db2da12fab06ba50b7da8476bae29","sha256:a40a0b98d6a75ec3dc881a07c99b31852fe9561a6c0f2b8cb30fa8b2fdbeac81"],"state_sha256":"11c524731a2b097c2a8778ef7f256dd21f2a1bf02770853979eab49ca811faaa"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kLJ8hGANoB0wLKFzbO1BM2wxxovRaPD3O0BHGlj5wsbdiKj0R0dWg2RVqSooW4JnWj2BLoglmdfBZX3Jd3oRAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T09:17:36.751118Z","bundle_sha256":"4f7d619babf5af9d47829689c9bf0ca5ba281b82ab3a288adc8184b43d989c45"}}