{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:PYTE46OLFG2QPPOQRLIWYISKLN","short_pith_number":"pith:PYTE46OL","canonical_record":{"source":{"id":"1711.09404","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-11-26T15:20:46Z","cross_cats_sorted":["cs.CR","cs.CV"],"title_canon_sha256":"b8d8198924e1f75831881261031b8f8897989163bbe93a5c9b95a40d72f0f2b3","abstract_canon_sha256":"ac73956741cfc41cf65fd7898c934a8e69534a2f78c35119205af705204da26c"},"schema_version":"1.0"},"canonical_sha256":"7e264e79cb29b507bdd08ad16c224a5b5cb408797b133489aea868f017f42906","source":{"kind":"arxiv","id":"1711.09404","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1711.09404","created_at":"2026-05-18T00:29:35Z"},{"alias_kind":"arxiv_version","alias_value":"1711.09404v1","created_at":"2026-05-18T00:29:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1711.09404","created_at":"2026-05-18T00:29:35Z"},{"alias_kind":"pith_short_12","alias_value":"PYTE46OLFG2Q","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"PYTE46OLFG2QPPOQ","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"PYTE46OL","created_at":"2026-05-18T12:31:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:PYTE46OLFG2QPPOQRLIWYISKLN","target":"record","payload":{"canonical_record":{"source":{"id":"1711.09404","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-11-26T15:20:46Z","cross_cats_sorted":["cs.CR","cs.CV"],"title_canon_sha256":"b8d8198924e1f75831881261031b8f8897989163bbe93a5c9b95a40d72f0f2b3","abstract_canon_sha256":"ac73956741cfc41cf65fd7898c934a8e69534a2f78c35119205af705204da26c"},"schema_version":"1.0"},"canonical_sha256":"7e264e79cb29b507bdd08ad16c224a5b5cb408797b133489aea868f017f42906","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:29:35.677847Z","signature_b64":"zH8g197QZdYQ57FGxZbMT4oLVKIN3IlQW/YEoJUBRWPVk8O1K8uZQ2Ngk4PrA4NqRdCattWiHsZ+Zh0TMYzDBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e264e79cb29b507bdd08ad16c224a5b5cb408797b133489aea868f017f42906","last_reissued_at":"2026-05-18T00:29:35.677143Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:29:35.677143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1711.09404","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":"2026-05-18T00:29:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rDYvU0HajMlRVC+CAS2rWep0JKyxD+T/JvHXmSZK7/0fchT0hReEClqlpwDSzlKlzBKfXndVSGyT8a0MDgZIBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T00:47:59.975402Z"},"content_sha256":"627e4be138a0ed752a0d3c742110a11d0d8a0cf1604d65e1546e3464402ade12","schema_version":"1.0","event_id":"sha256:627e4be138a0ed752a0d3c742110a11d0d8a0cf1604d65e1546e3464402ade12"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:PYTE46OLFG2QPPOQRLIWYISKLN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.CV"],"primary_cat":"cs.LG","authors_text":"Andrew Slavin Ross, Finale Doshi-Velez","submitted_at":"2017-11-26T15:20:46Z","abstract_excerpt":"Deep neural networks have proven remarkably effective at solving many classification problems, but have been criticized recently for two major weaknesses: the reasons behind their predictions are uninterpretable, and the predictions themselves can often be fooled by small adversarial perturbations. These problems pose major obstacles for the adoption of neural networks in domains that require security or transparency. In this work, we evaluate the effectiveness of defenses that differentiably penalize the degree to which small changes in inputs can alter model predictions. Across multiple atta"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1711.09404","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":"2026-05-18T00:29:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UCvk4DHLOrETJ1J1wbJYN+pozTpsx8wYA35pJzI4oiqGzA0C0VitSUuF8Yv0JN2WtX7YWz49r5IIEGUrh7GTDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T00:47:59.976098Z"},"content_sha256":"778f2988686967b20ff44830615f1c49307e5a1be629baa27207dd44d42f0c57","schema_version":"1.0","event_id":"sha256:778f2988686967b20ff44830615f1c49307e5a1be629baa27207dd44d42f0c57"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PYTE46OLFG2QPPOQRLIWYISKLN/bundle.json","state_url":"https://pith.science/pith/PYTE46OLFG2QPPOQRLIWYISKLN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PYTE46OLFG2QPPOQRLIWYISKLN/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":"2026-05-26T00:47:59Z","links":{"resolver":"https://pith.science/pith/PYTE46OLFG2QPPOQRLIWYISKLN","bundle":"https://pith.science/pith/PYTE46OLFG2QPPOQRLIWYISKLN/bundle.json","state":"https://pith.science/pith/PYTE46OLFG2QPPOQRLIWYISKLN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PYTE46OLFG2QPPOQRLIWYISKLN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:PYTE46OLFG2QPPOQRLIWYISKLN","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":"ac73956741cfc41cf65fd7898c934a8e69534a2f78c35119205af705204da26c","cross_cats_sorted":["cs.CR","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-11-26T15:20:46Z","title_canon_sha256":"b8d8198924e1f75831881261031b8f8897989163bbe93a5c9b95a40d72f0f2b3"},"schema_version":"1.0","source":{"id":"1711.09404","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1711.09404","created_at":"2026-05-18T00:29:35Z"},{"alias_kind":"arxiv_version","alias_value":"1711.09404v1","created_at":"2026-05-18T00:29:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1711.09404","created_at":"2026-05-18T00:29:35Z"},{"alias_kind":"pith_short_12","alias_value":"PYTE46OLFG2Q","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"PYTE46OLFG2QPPOQ","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"PYTE46OL","created_at":"2026-05-18T12:31:37Z"}],"graph_snapshots":[{"event_id":"sha256:778f2988686967b20ff44830615f1c49307e5a1be629baa27207dd44d42f0c57","target":"graph","created_at":"2026-05-18T00:29:35Z","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 have proven remarkably effective at solving many classification problems, but have been criticized recently for two major weaknesses: the reasons behind their predictions are uninterpretable, and the predictions themselves can often be fooled by small adversarial perturbations. These problems pose major obstacles for the adoption of neural networks in domains that require security or transparency. In this work, we evaluate the effectiveness of defenses that differentiably penalize the degree to which small changes in inputs can alter model predictions. Across multiple atta","authors_text":"Andrew Slavin Ross, Finale Doshi-Velez","cross_cats":["cs.CR","cs.CV"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-11-26T15:20:46Z","title":"Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1711.09404","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":"sha256:627e4be138a0ed752a0d3c742110a11d0d8a0cf1604d65e1546e3464402ade12","target":"record","created_at":"2026-05-18T00:29:35Z","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":"ac73956741cfc41cf65fd7898c934a8e69534a2f78c35119205af705204da26c","cross_cats_sorted":["cs.CR","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-11-26T15:20:46Z","title_canon_sha256":"b8d8198924e1f75831881261031b8f8897989163bbe93a5c9b95a40d72f0f2b3"},"schema_version":"1.0","source":{"id":"1711.09404","kind":"arxiv","version":1}},"canonical_sha256":"7e264e79cb29b507bdd08ad16c224a5b5cb408797b133489aea868f017f42906","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7e264e79cb29b507bdd08ad16c224a5b5cb408797b133489aea868f017f42906","first_computed_at":"2026-05-18T00:29:35.677143Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:29:35.677143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zH8g197QZdYQ57FGxZbMT4oLVKIN3IlQW/YEoJUBRWPVk8O1K8uZQ2Ngk4PrA4NqRdCattWiHsZ+Zh0TMYzDBw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:29:35.677847Z","signed_message":"canonical_sha256_bytes"},"source_id":"1711.09404","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:627e4be138a0ed752a0d3c742110a11d0d8a0cf1604d65e1546e3464402ade12","sha256:778f2988686967b20ff44830615f1c49307e5a1be629baa27207dd44d42f0c57"],"state_sha256":"874dd01b1c84895be4bb7f0bdf48b9c1d0091f195dc292ff57fa6b78cca05ea2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IR6HYFajHFlAAFqHXMjSNitrE4MX8JaLQR6fAzD0n3o4Zp4MYo/ZFnQCdPJ341SrFQo7zBMiIjY9zkmDJMKeBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T00:47:59.980224Z","bundle_sha256":"8f2417b815cda26fcaf682e6e1dbd9608d87f8ab4caaa784eb7f6797d20ddf82"}}