{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:4WAJ3FY2SURQ5PFJ7U7RD4VFWT","short_pith_number":"pith:4WAJ3FY2","canonical_record":{"source":{"id":"1906.03499","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T18:36:16Z","cross_cats_sorted":["cs.CR","stat.ML"],"title_canon_sha256":"3a2b4795b87c47d2933e02c59baca59b28a13ab7e1ae358e638ed8e65c4e15cf","abstract_canon_sha256":"024cc74b44bc6630501ac87bff8466c93b6b9f0caa2b5eecd01ddb1a9b7e6ffa"},"schema_version":"1.0"},"canonical_sha256":"e5809d971a95230ebca9fd3f11f2a5b4f153b314f385d68452bf38b97405cec5","source":{"kind":"arxiv","id":"1906.03499","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.03499","created_at":"2026-05-17T23:43:48Z"},{"alias_kind":"arxiv_version","alias_value":"1906.03499v1","created_at":"2026-05-17T23:43:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.03499","created_at":"2026-05-17T23:43:48Z"},{"alias_kind":"pith_short_12","alias_value":"4WAJ3FY2SURQ","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_16","alias_value":"4WAJ3FY2SURQ5PFJ","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_8","alias_value":"4WAJ3FY2","created_at":"2026-05-18T12:33:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:4WAJ3FY2SURQ5PFJ7U7RD4VFWT","target":"record","payload":{"canonical_record":{"source":{"id":"1906.03499","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T18:36:16Z","cross_cats_sorted":["cs.CR","stat.ML"],"title_canon_sha256":"3a2b4795b87c47d2933e02c59baca59b28a13ab7e1ae358e638ed8e65c4e15cf","abstract_canon_sha256":"024cc74b44bc6630501ac87bff8466c93b6b9f0caa2b5eecd01ddb1a9b7e6ffa"},"schema_version":"1.0"},"canonical_sha256":"e5809d971a95230ebca9fd3f11f2a5b4f153b314f385d68452bf38b97405cec5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:48.275421Z","signature_b64":"wvdPfDWhPuKitGpQYkV+/ZK8S8pjkrx+PmsoJ6+aLrD+JduwbcLwRF/ZLHbfX5+pM9po5VlBL8o4gRgQfxZjBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5809d971a95230ebca9fd3f11f2a5b4f153b314f385d68452bf38b97405cec5","last_reissued_at":"2026-05-17T23:43:48.274873Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:48.274873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1906.03499","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":"2026-05-17T23:43:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CkzI/Zqmmrp9lTpoK0besTdfSJ1L7IINRp8++fBZa2v1DdVUVQ+C14RKClkMIHVrOVAAcrNdREgThawAiFDAAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T06:20:29.954806Z"},"content_sha256":"066ebbd8b53635441a41dc1be5d64398dd1d17dde504fd73796570a491abeb44","schema_version":"1.0","event_id":"sha256:066ebbd8b53635441a41dc1be5d64398dd1d17dde504fd73796570a491abeb44"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:4WAJ3FY2SURQ5PFJ7U7RD4VFWT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ML-LOO: Detecting Adversarial Examples with Feature Attribution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Cho-Jui Hsieh, Jane-Ling Wang, Jianbo Chen, Michael I. Jordan, Puyudi Yang","submitted_at":"2019-06-08T18:36:16Z","abstract_excerpt":"Deep neural networks obtain state-of-the-art performance on a series of tasks. However, they are easily fooled by adding a small adversarial perturbation to input. The perturbation is often human imperceptible on image data. We observe a significant difference in feature attributions of adversarially crafted examples from those of original ones. Based on this observation, we introduce a new framework to detect adversarial examples through thresholding a scale estimate of feature attribution scores. Furthermore, we extend our method to include multi-layer feature attributions in order to tackle"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.03499","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":""},"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":"2026-05-17T23:43:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AXYF8WP7HK3A4+xsz7V56kS++KWj3d/AU4hWKTgfGd04JYDTyI2E+IE5LXN5F5BZxgr+qGxE7v1lXHb2yiNmCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T06:20:29.955177Z"},"content_sha256":"5f123a9866a058881c6ba20a80a7b475cc555016f06b02088e83eccc668b8501","schema_version":"1.0","event_id":"sha256:5f123a9866a058881c6ba20a80a7b475cc555016f06b02088e83eccc668b8501"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4WAJ3FY2SURQ5PFJ7U7RD4VFWT/bundle.json","state_url":"https://pith.science/pith/4WAJ3FY2SURQ5PFJ7U7RD4VFWT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4WAJ3FY2SURQ5PFJ7U7RD4VFWT/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":"2026-07-04T06:20:29Z","links":{"resolver":"https://pith.science/pith/4WAJ3FY2SURQ5PFJ7U7RD4VFWT","bundle":"https://pith.science/pith/4WAJ3FY2SURQ5PFJ7U7RD4VFWT/bundle.json","state":"https://pith.science/pith/4WAJ3FY2SURQ5PFJ7U7RD4VFWT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4WAJ3FY2SURQ5PFJ7U7RD4VFWT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:4WAJ3FY2SURQ5PFJ7U7RD4VFWT","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":"024cc74b44bc6630501ac87bff8466c93b6b9f0caa2b5eecd01ddb1a9b7e6ffa","cross_cats_sorted":["cs.CR","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T18:36:16Z","title_canon_sha256":"3a2b4795b87c47d2933e02c59baca59b28a13ab7e1ae358e638ed8e65c4e15cf"},"schema_version":"1.0","source":{"id":"1906.03499","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.03499","created_at":"2026-05-17T23:43:48Z"},{"alias_kind":"arxiv_version","alias_value":"1906.03499v1","created_at":"2026-05-17T23:43:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.03499","created_at":"2026-05-17T23:43:48Z"},{"alias_kind":"pith_short_12","alias_value":"4WAJ3FY2SURQ","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_16","alias_value":"4WAJ3FY2SURQ5PFJ","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_8","alias_value":"4WAJ3FY2","created_at":"2026-05-18T12:33:10Z"}],"graph_snapshots":[{"event_id":"sha256:5f123a9866a058881c6ba20a80a7b475cc555016f06b02088e83eccc668b8501","target":"graph","created_at":"2026-05-17T23:43:48Z","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"},"paper":{"abstract_excerpt":"Deep neural networks obtain state-of-the-art performance on a series of tasks. However, they are easily fooled by adding a small adversarial perturbation to input. The perturbation is often human imperceptible on image data. We observe a significant difference in feature attributions of adversarially crafted examples from those of original ones. Based on this observation, we introduce a new framework to detect adversarial examples through thresholding a scale estimate of feature attribution scores. Furthermore, we extend our method to include multi-layer feature attributions in order to tackle","authors_text":"Cho-Jui Hsieh, Jane-Ling Wang, Jianbo Chen, Michael I. Jordan, Puyudi Yang","cross_cats":["cs.CR","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T18:36:16Z","title":"ML-LOO: Detecting Adversarial Examples with Feature Attribution"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.03499","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":"sha256:066ebbd8b53635441a41dc1be5d64398dd1d17dde504fd73796570a491abeb44","target":"record","created_at":"2026-05-17T23:43:48Z","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":"024cc74b44bc6630501ac87bff8466c93b6b9f0caa2b5eecd01ddb1a9b7e6ffa","cross_cats_sorted":["cs.CR","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T18:36:16Z","title_canon_sha256":"3a2b4795b87c47d2933e02c59baca59b28a13ab7e1ae358e638ed8e65c4e15cf"},"schema_version":"1.0","source":{"id":"1906.03499","kind":"arxiv","version":1}},"canonical_sha256":"e5809d971a95230ebca9fd3f11f2a5b4f153b314f385d68452bf38b97405cec5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e5809d971a95230ebca9fd3f11f2a5b4f153b314f385d68452bf38b97405cec5","first_computed_at":"2026-05-17T23:43:48.274873Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:43:48.274873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wvdPfDWhPuKitGpQYkV+/ZK8S8pjkrx+PmsoJ6+aLrD+JduwbcLwRF/ZLHbfX5+pM9po5VlBL8o4gRgQfxZjBg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:43:48.275421Z","signed_message":"canonical_sha256_bytes"},"source_id":"1906.03499","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:066ebbd8b53635441a41dc1be5d64398dd1d17dde504fd73796570a491abeb44","sha256:5f123a9866a058881c6ba20a80a7b475cc555016f06b02088e83eccc668b8501"],"state_sha256":"7b767722b221b262fa49d0e3698791a412b81b9239e8939efc9a53785c715510"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VP2Tq+1E7Ag4m3o0mCE98X4KDEnwAgaq+p3Ea0sumdvw8YTSvdj49fAyoL2cVOWqSXwDYtz1qs4jh5B5zowADw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T06:20:29.957216Z","bundle_sha256":"d2131b108f47220339d0701d632d79c264221775ce8885b58393297eb409aa06"}}