{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:WCAGZL5XRSTYMA34W4EEOS5XNV","short_pith_number":"pith:WCAGZL5X","canonical_record":{"source":{"id":"2605.12792","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T22:10:01Z","cross_cats_sorted":[],"title_canon_sha256":"cc144459d4428b17683fb9be085960a0f80d10476b75d2cd57182e36e9ed3cf5","abstract_canon_sha256":"ae02a9f9cefab6bbac9580f19312798456b5979c888816dad366bf7511c5135c"},"schema_version":"1.0"},"canonical_sha256":"b0806cafb78ca786037cb708474bb76d4542e2e689f7703f7c76a4e6fcfa03c1","source":{"kind":"arxiv","id":"2605.12792","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12792","created_at":"2026-05-18T03:09:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12792v1","created_at":"2026-05-18T03:09:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12792","created_at":"2026-05-18T03:09:12Z"},{"alias_kind":"pith_short_12","alias_value":"WCAGZL5XRSTY","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"WCAGZL5XRSTYMA34","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"WCAGZL5X","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:WCAGZL5XRSTYMA34W4EEOS5XNV","target":"record","payload":{"canonical_record":{"source":{"id":"2605.12792","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T22:10:01Z","cross_cats_sorted":[],"title_canon_sha256":"cc144459d4428b17683fb9be085960a0f80d10476b75d2cd57182e36e9ed3cf5","abstract_canon_sha256":"ae02a9f9cefab6bbac9580f19312798456b5979c888816dad366bf7511c5135c"},"schema_version":"1.0"},"canonical_sha256":"b0806cafb78ca786037cb708474bb76d4542e2e689f7703f7c76a4e6fcfa03c1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:12.885092Z","signature_b64":"h0BQ8JnP9EWcCEYNyomM/VoBCBk6Uc8bS1GM1TahOJ71epEVJSgbETmCGKI3MCm6y4wklFnKXfF4/Ccku3g3Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0806cafb78ca786037cb708474bb76d4542e2e689f7703f7c76a4e6fcfa03c1","last_reissued_at":"2026-05-18T03:09:12.884625Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:12.884625Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.12792","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":"2026-05-18T03:09:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zpqZ+X29RGKhqOIEtb9aNXl7WNGSrubmHdpq44jIdyURFb3bBSP1a1I3JweVg0HFQh+M85M0o3X5ETmCAC4jDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T11:12:58.169832Z"},"content_sha256":"59d9a45c5e5dfa87c86ebf25d20fd42e61dad0b52483ea8f108673f0ac8abd10","schema_version":"1.0","event_id":"sha256:59d9a45c5e5dfa87c86ebf25d20fd42e61dad0b52483ea8f108673f0ac8abd10"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:WCAGZL5XRSTYMA34W4EEOS5XNV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SoK: A Comprehensive Analysis of the Current Status of Neural Tangent Generalization Attacks with Research Directions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"NTGA, the first black-box clean-label generalization attack, is vulnerable to adversarial training and image transformations while newer attacks perform better.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Kaiqi Xiong, Thushari Hapuarachchi","submitted_at":"2026-05-12T22:10:01Z","abstract_excerpt":"There is recently a serious issue that Deep Neural Networks (DNNs) training uses more and more unauthorized data. A clean-label generalization attack, one type of data poisoning attacks, has been suggested to address this issue. The Neural Tangent Generalization Attack (NTGA) is considered as the first well-known clean-label generalization attack under the black-box settings, which provided an unprecedented step in data protection approaches. In this paper, we conduct a comprehensive analysis on the state-of-the-art of NTGA; to the best of our knowledge, this is the first thorough analysis reg"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"NTGA is the first well-known clean-label generalization attack under the black-box settings, yet it is vulnerable to adversarial training and image transformations, and several recently proposed clean-label generalization attacks outperform NTGA on data protection.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the authors' own experiments are representative enough to establish general vulnerabilities and performance comparisons across models, datasets, and threat models without full methodological details provided.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"NTGA is the first clean-label generalization attack under black-box settings but is vulnerable to adversarial training and image transformations, with newer attacks outperforming it.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"NTGA, the first black-box clean-label generalization attack, is vulnerable to adversarial training and image transformations while newer attacks perform better.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d5999aa1a9b62f69eca26e40cec102a1caadc8122dec1208a28dc15a30c95285"},"source":{"id":"2605.12792","kind":"arxiv","version":1},"verdict":{"id":"b5033888-d87d-4a1d-a567-1a4a3d3d259e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T20:38:05.432423Z","strongest_claim":"NTGA is the first well-known clean-label generalization attack under the black-box settings, yet it is vulnerable to adversarial training and image transformations, and several recently proposed clean-label generalization attacks outperform NTGA on data protection.","one_line_summary":"NTGA is the first clean-label generalization attack under black-box settings but is vulnerable to adversarial training and image transformations, with newer attacks outperforming it.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the authors' own experiments are representative enough to establish general vulnerabilities and performance comparisons across models, datasets, and threat models without full methodological details provided.","pith_extraction_headline":"NTGA, the first black-box clean-label generalization attack, is vulnerable to adversarial training and image transformations while newer attacks perform better."},"references":{"count":112,"sample":[{"doi":"","year":null,"title":"d.].Google Cloud AutoML","work_id":"8c854138-9886-4645-b1a2-edeb44e96802","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"d.].Google Cloud Vision API","work_id":"6f8b0a67-012a-4828-a24e-8006cb94a262","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"d.].Machine Learning on A WS","work_id":"2df67c63-b7f0-4669-a6e3-a7ca5ce7d9d2","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"d.].Use a Color Matrix to Transform a Single Color","work_id":"93e1270d-28fa-4dab-b7ee-73629a6dac01","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Ibrahim M Ahmed and Manar Younis Kashmoola. 2021. Threats on Machine Learning Technique by Data Poisoning Attack: A Survey. InInternational Con- ference on Advances in Cyber Security. Springer, 586–60","work_id":"35deb06e-c410-4e0e-9cdb-d0c522dcd787","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":112,"snapshot_sha256":"8d6803fff45dc8836a4f0d54dd046ad857bfbe28221514dcc4be3cfefdcdb2b6","internal_anchors":14},"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":"b5033888-d87d-4a1d-a567-1a4a3d3d259e"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VUJq/2GZeyOomlAsiAiRUXcrv6ma/jp2YSz1MQZixat+pv8MP/gg9Geh5lKZX5alzEqWdu3LlS2hqrziCnC+BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T11:12:58.170508Z"},"content_sha256":"fdf75f991b12de103f2a0c6fecaa7fa974c7ea8d458344cea541e2d4ba0a7677","schema_version":"1.0","event_id":"sha256:fdf75f991b12de103f2a0c6fecaa7fa974c7ea8d458344cea541e2d4ba0a7677"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:WCAGZL5XRSTYMA34W4EEOS5XNV","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1016/j.dcan.SoK) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Chen Wang, Jian Chen, Yang Yang, Xiaoqiang Ma, and Jiangchuan Liu. 2022. Poisoning attacks and countermeasures in intelligent networks: Status quo and prospects.Digit. Commun. Networks8, 2 (2022), 225–234. doi:10.1016/j.dcan. SoK: A Compreh","arxiv_id":"2605.12792","detector":"doi_compliance","evidence":{"ref_index":95,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Chen Wang, Jian Chen, Yang Yang, Xiaoqiang Ma, and Jiangchuan Liu. 2022. Poisoning attacks and countermeasures in intelligent networks: Status quo and prospects.Digit. Commun. Networks8, 2 (2022), 225–234. doi:10.1016/j.dcan. SoK: A Compreh","reconstructed_doi":"10.1016/j.dcan.SoK"},"severity":"advisory","ref_index":95,"audited_at":"2026-05-19T07:08:19.926242Z","event_type":"pith.integrity.v1","detected_doi":"10.1016/j.dcan.SoK","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"76171e6461cc68687b9e2e7c198d3dfa37af6779e081f60ed57bf5d0d49e7d52","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":264,"payload_sha256":"9e0e4cb8971b287f67489055948571d1708b4569ef8c5b37a5c16d696d2d109c","signature_b64":"m/g1OG26LllV4Uhq2r3QqzD7PLEbaQezp4XLsSyguYViQtLfg+MBX8tjggJLUFjIX+UDxJgFklGs2+LbK9ZMDQ==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-19T07:11:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4X91AeGkPTwtp3ttcdB9Jaco1uLJqLBEZNCdlvzx6bldiC1VTqaFLhdAEAmGVAQPlOWZN89M3cAj0DCUAQE8DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T11:12:58.171410Z"},"content_sha256":"09aeed0621bdfa165eeef7d7d735f118a2c38c6ee95f3021fc5a2161d1597b8d","schema_version":"1.0","event_id":"sha256:09aeed0621bdfa165eeef7d7d735f118a2c38c6ee95f3021fc5a2161d1597b8d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WCAGZL5XRSTYMA34W4EEOS5XNV/bundle.json","state_url":"https://pith.science/pith/WCAGZL5XRSTYMA34W4EEOS5XNV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WCAGZL5XRSTYMA34W4EEOS5XNV/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":"2026-05-21T11:12:58Z","links":{"resolver":"https://pith.science/pith/WCAGZL5XRSTYMA34W4EEOS5XNV","bundle":"https://pith.science/pith/WCAGZL5XRSTYMA34W4EEOS5XNV/bundle.json","state":"https://pith.science/pith/WCAGZL5XRSTYMA34W4EEOS5XNV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WCAGZL5XRSTYMA34W4EEOS5XNV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WCAGZL5XRSTYMA34W4EEOS5XNV","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"ae02a9f9cefab6bbac9580f19312798456b5979c888816dad366bf7511c5135c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T22:10:01Z","title_canon_sha256":"cc144459d4428b17683fb9be085960a0f80d10476b75d2cd57182e36e9ed3cf5"},"schema_version":"1.0","source":{"id":"2605.12792","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12792","created_at":"2026-05-18T03:09:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12792v1","created_at":"2026-05-18T03:09:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12792","created_at":"2026-05-18T03:09:12Z"},{"alias_kind":"pith_short_12","alias_value":"WCAGZL5XRSTY","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"WCAGZL5XRSTYMA34","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"WCAGZL5X","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:fdf75f991b12de103f2a0c6fecaa7fa974c7ea8d458344cea541e2d4ba0a7677","target":"graph","created_at":"2026-05-18T03:09:12Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"NTGA is the first well-known clean-label generalization attack under the black-box settings, yet it is vulnerable to adversarial training and image transformations, and several recently proposed clean-label generalization attacks outperform NTGA on data protection."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the authors' own experiments are representative enough to establish general vulnerabilities and performance comparisons across models, datasets, and threat models without full methodological details provided."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"NTGA is the first clean-label generalization attack under black-box settings but is vulnerable to adversarial training and image transformations, with newer attacks outperforming it."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"NTGA, the first black-box clean-label generalization attack, is vulnerable to adversarial training and image transformations while newer attacks perform better."}],"snapshot_sha256":"d5999aa1a9b62f69eca26e40cec102a1caadc8122dec1208a28dc15a30c95285"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"There is recently a serious issue that Deep Neural Networks (DNNs) training uses more and more unauthorized data. A clean-label generalization attack, one type of data poisoning attacks, has been suggested to address this issue. The Neural Tangent Generalization Attack (NTGA) is considered as the first well-known clean-label generalization attack under the black-box settings, which provided an unprecedented step in data protection approaches. In this paper, we conduct a comprehensive analysis on the state-of-the-art of NTGA; to the best of our knowledge, this is the first thorough analysis reg","authors_text":"Kaiqi Xiong, Thushari Hapuarachchi","cross_cats":[],"headline":"NTGA, the first black-box clean-label generalization attack, is vulnerable to adversarial training and image transformations while newer attacks perform better.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T22:10:01Z","title":"SoK: A Comprehensive Analysis of the Current Status of Neural Tangent Generalization Attacks with Research Directions"},"references":{"count":112,"internal_anchors":14,"resolved_work":112,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"d.].Google Cloud AutoML","work_id":"8c854138-9886-4645-b1a2-edeb44e96802","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"d.].Google Cloud Vision API","work_id":"6f8b0a67-012a-4828-a24e-8006cb94a262","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"d.].Machine Learning on A WS","work_id":"2df67c63-b7f0-4669-a6e3-a7ca5ce7d9d2","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"d.].Use a Color Matrix to Transform a Single Color","work_id":"93e1270d-28fa-4dab-b7ee-73629a6dac01","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Ibrahim M Ahmed and Manar Younis Kashmoola. 2021. Threats on Machine Learning Technique by Data Poisoning Attack: A Survey. InInternational Con- ference on Advances in Cyber Security. Springer, 586–60","work_id":"35deb06e-c410-4e0e-9cdb-d0c522dcd787","year":2021}],"snapshot_sha256":"8d6803fff45dc8836a4f0d54dd046ad857bfbe28221514dcc4be3cfefdcdb2b6"},"source":{"id":"2605.12792","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T20:38:05.432423Z","id":"b5033888-d87d-4a1d-a567-1a4a3d3d259e","model_set":{"reader":"grok-4.3"},"one_line_summary":"NTGA is the first clean-label generalization attack under black-box settings but is vulnerable to adversarial training and image transformations, with newer attacks outperforming it.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"NTGA, the first black-box clean-label generalization attack, is vulnerable to adversarial training and image transformations while newer attacks perform better.","strongest_claim":"NTGA is the first well-known clean-label generalization attack under the black-box settings, yet it is vulnerable to adversarial training and image transformations, and several recently proposed clean-label generalization attacks outperform NTGA on data protection.","weakest_assumption":"That the authors' own experiments are representative enough to establish general vulnerabilities and performance comparisons across models, datasets, and threat models without full methodological details provided."}},"verdict_id":"b5033888-d87d-4a1d-a567-1a4a3d3d259e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:59d9a45c5e5dfa87c86ebf25d20fd42e61dad0b52483ea8f108673f0ac8abd10","target":"record","created_at":"2026-05-18T03:09:12Z","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":"ae02a9f9cefab6bbac9580f19312798456b5979c888816dad366bf7511c5135c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T22:10:01Z","title_canon_sha256":"cc144459d4428b17683fb9be085960a0f80d10476b75d2cd57182e36e9ed3cf5"},"schema_version":"1.0","source":{"id":"2605.12792","kind":"arxiv","version":1}},"canonical_sha256":"b0806cafb78ca786037cb708474bb76d4542e2e689f7703f7c76a4e6fcfa03c1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b0806cafb78ca786037cb708474bb76d4542e2e689f7703f7c76a4e6fcfa03c1","first_computed_at":"2026-05-18T03:09:12.884625Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:12.884625Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"h0BQ8JnP9EWcCEYNyomM/VoBCBk6Uc8bS1GM1TahOJ71epEVJSgbETmCGKI3MCm6y4wklFnKXfF4/Ccku3g3Cg==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:12.885092Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12792","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:59d9a45c5e5dfa87c86ebf25d20fd42e61dad0b52483ea8f108673f0ac8abd10","sha256:fdf75f991b12de103f2a0c6fecaa7fa974c7ea8d458344cea541e2d4ba0a7677","sha256:09aeed0621bdfa165eeef7d7d735f118a2c38c6ee95f3021fc5a2161d1597b8d"],"state_sha256":"e75ef883f3883728a013af9908241dea055fe64e823bdc6d1ea27c1b75b48c1c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XYYIN5Oh1dslTajRUHEXQ0xsw5eYs1HHbysIESDpsW5johQmha1UVx5ZGKAKa1puXNmquAI6/U2wy63Qb2d1Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-21T11:12:58.173769Z","bundle_sha256":"d383db5e317ebba7d23ef54ea4b4003c24de6a4cf81dcde64eecf47f614bd1ea"}}