{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MU5Q3RHHUXYZLVLL4VNLVB6TJM","short_pith_number":"pith:MU5Q3RHH","canonical_record":{"source":{"id":"2605.15688","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-05-15T07:22:59Z","cross_cats_sorted":["cs.AI","cs.LG","math.PR"],"title_canon_sha256":"8d586ce140c497c4ae3bba821a7b9172e1272da09adfe3fcac59fdfce40bfc74","abstract_canon_sha256":"a941485d7933e4d858d2edede58b35dc959ff72cc3d1707b6cdf078f35c9902a"},"schema_version":"1.0"},"canonical_sha256":"653b0dc4e7a5f195d56be55aba87d34b128ff42f9a21de9e1c32d6a9a41f87db","source":{"kind":"arxiv","id":"2605.15688","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15688","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15688v1","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15688","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"pith_short_12","alias_value":"MU5Q3RHHUXYZ","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"pith_short_16","alias_value":"MU5Q3RHHUXYZLVLL","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"pith_short_8","alias_value":"MU5Q3RHH","created_at":"2026-05-20T00:01:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MU5Q3RHHUXYZLVLL4VNLVB6TJM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.15688","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-05-15T07:22:59Z","cross_cats_sorted":["cs.AI","cs.LG","math.PR"],"title_canon_sha256":"8d586ce140c497c4ae3bba821a7b9172e1272da09adfe3fcac59fdfce40bfc74","abstract_canon_sha256":"a941485d7933e4d858d2edede58b35dc959ff72cc3d1707b6cdf078f35c9902a"},"schema_version":"1.0"},"canonical_sha256":"653b0dc4e7a5f195d56be55aba87d34b128ff42f9a21de9e1c32d6a9a41f87db","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:12.527520Z","signature_b64":"vmEAFVJZ0NhymvIUjps8YDmgPWPVqcECemDSWb55hNOvJ5k9pqD17/vLWLG/eAuD7lD+bILuCb4Pt6phlHAdAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"653b0dc4e7a5f195d56be55aba87d34b128ff42f9a21de9e1c32d6a9a41f87db","last_reissued_at":"2026-05-20T00:01:12.526593Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:12.526593Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.15688","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":"2026-05-20T00:01:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lnSeU+CxLQK/C7lUOtbndihr5dG0k3+9wxNUpc0GYROz9e1zhjCpN3ZbgYOEO91BmRBOXK8wAA3jqQ0WksaxAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T23:18:27.369861Z"},"content_sha256":"79ee46ac601a2627307856bed24619bf5ba96e2bf0e88cf082d0e12ced7e07cb","schema_version":"1.0","event_id":"sha256:79ee46ac601a2627307856bed24619bf5ba96e2bf0e88cf082d0e12ced7e07cb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MU5Q3RHHUXYZLVLL4VNLVB6TJM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"$\\alpha$-TCAV: A Unified Framework for Testing with Concept Activation Vectors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"The standard TCAV score has non-decaying variance from its discontinuous indicator, which α-TCAV fixes by substituting a parameterized smooth function for a unified probabilistic framework.","cross_cats":["cs.AI","cs.LG","math.PR"],"primary_cat":"stat.ML","authors_text":"Alexander Jung, Ekkehard Schnoor, Jawher Said, Malik Tiomoko, Wojciech Samek","submitted_at":"2026-05-15T07:22:59Z","abstract_excerpt":"Concept Activation Vectors (CAVs) are a fundamental tool for concept-based explainability in deep learning, yet their practical utility is limited by statistical instability. We analyze the stochastic nature of CAVs and the Testing with CAVs (TCAV) method, deriving the distributions of major CAV classes including PatternCAV, FastCAV, and ridge regression-based CAVs. We then identify a fundamental flaw in the standard TCAV score: its reliance on a discontinuous indicator function induces non-decaying variance in critical regimes. To address this, we introduce $\\alpha$-TCAV, a generalized framew"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We identify a fundamental flaw in the standard TCAV score: its reliance on a discontinuous indicator function induces non-decaying variance in critical regimes. To address this, we introduce α-TCAV, a generalized framework that replaces the indicator with a parameterized smooth function, yielding a unified probabilistic formulation that subsumes both TCAV and Multi-TCAV.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The derivations of the distributions for PatternCAV, FastCAV, and ridge-regression CAVs are analytically tractable and that the smooth function in α-TCAV can be tuned to achieve either imitation of Multi-TCAV or a calibrated Bayes-optimal measure without introducing new bias. This premise enters in the stochastic analysis section and the guidance on tuning the parameter.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"α-TCAV replaces TCAV's hard indicator with a tunable smooth function to create a unified probabilistic framework with lower variance and guidance for parameter choice or Bayes-optimal scoring.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The standard TCAV score has non-decaying variance from its discontinuous indicator, which α-TCAV fixes by substituting a parameterized smooth function for a unified probabilistic framework.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"477dbfda13a66c7a4a0e37f271115698b415d418ec56aee7cca43c0279d783a7"},"source":{"id":"2605.15688","kind":"arxiv","version":1},"verdict":{"id":"1a27c302-944f-4e44-a65a-0fcc8687b08d","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T19:44:12.106838Z","strongest_claim":"We identify a fundamental flaw in the standard TCAV score: its reliance on a discontinuous indicator function induces non-decaying variance in critical regimes. To address this, we introduce α-TCAV, a generalized framework that replaces the indicator with a parameterized smooth function, yielding a unified probabilistic formulation that subsumes both TCAV and Multi-TCAV.","one_line_summary":"α-TCAV replaces TCAV's hard indicator with a tunable smooth function to create a unified probabilistic framework with lower variance and guidance for parameter choice or Bayes-optimal scoring.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The derivations of the distributions for PatternCAV, FastCAV, and ridge-regression CAVs are analytically tractable and that the smooth function in α-TCAV can be tuned to achieve either imitation of Multi-TCAV or a calibrated Bayes-optimal measure without introducing new bias. This premise enters in the stochastic analysis section and the guidance on tuning the parameter.","pith_extraction_headline":"The standard TCAV score has non-decaying variance from its discontinuous indicator, which α-TCAV fixes by substituting a parameterized smooth function for a unified probabilistic framework."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15688/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T20:01:19.233329Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T19:51:29.209145Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T19:33:29.223945Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:21:56.044684Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"8b1a57cb22acc2942c59f0a29d2cdd716acfafd254b0159d827b4196ec2fdaeb"},"references":{"count":151,"sample":[{"doi":"","year":1905,"title":"Zur Elektrodynamik bewegter Körper","work_id":"f62dda56-8333-47e0-bc47-249c3fbb0cbe","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"The Annals of Applied Probability , volume=","work_id":"4393abc6-a331-4ace-a7b4-41b21e0bf3f5","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"The Thirteenth International Conference on Learning Representations , year=","work_id":"d465d843-18f0-4777-accf-6ce6e6dccefb","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Characterization of Gaussian Universality Breakdown in High-Dimensional Empirical Risk Minimization","work_id":"b7744d6f-6170-4ccf-a706-a023dc596384","ref_index":4,"cited_arxiv_id":"2604.03146","is_internal_anchor":true},{"doi":"","year":1993,"title":"The \\ Companion","work_id":"de8bc6e8-e1fa-4090-a818-24103cd763f9","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":151,"snapshot_sha256":"9e39aa12c286e40582d729de1960e9923b443772b61bd958d9497a40972a95a5","internal_anchors":4},"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":"1a27c302-944f-4e44-a65a-0fcc8687b08d"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:01:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JeiE4p8i+teBVRXrBdR1ylrcMyndI/X1Iw8q01ZIqbBfOjJrW02mXyVmJimF5kP/npIh7VL2jb52uPHXU2w2CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T23:18:27.371340Z"},"content_sha256":"c3019943fe182f2228cf26c00697b209def8687c59d55abf78e898cebe05216c","schema_version":"1.0","event_id":"sha256:c3019943fe182f2228cf26c00697b209def8687c59d55abf78e898cebe05216c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MU5Q3RHHUXYZLVLL4VNLVB6TJM/bundle.json","state_url":"https://pith.science/pith/MU5Q3RHHUXYZLVLL4VNLVB6TJM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MU5Q3RHHUXYZLVLL4VNLVB6TJM/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":"2026-05-25T23:18:27Z","links":{"resolver":"https://pith.science/pith/MU5Q3RHHUXYZLVLL4VNLVB6TJM","bundle":"https://pith.science/pith/MU5Q3RHHUXYZLVLL4VNLVB6TJM/bundle.json","state":"https://pith.science/pith/MU5Q3RHHUXYZLVLL4VNLVB6TJM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MU5Q3RHHUXYZLVLL4VNLVB6TJM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MU5Q3RHHUXYZLVLL4VNLVB6TJM","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":"a941485d7933e4d858d2edede58b35dc959ff72cc3d1707b6cdf078f35c9902a","cross_cats_sorted":["cs.AI","cs.LG","math.PR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-05-15T07:22:59Z","title_canon_sha256":"8d586ce140c497c4ae3bba821a7b9172e1272da09adfe3fcac59fdfce40bfc74"},"schema_version":"1.0","source":{"id":"2605.15688","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15688","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15688v1","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15688","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"pith_short_12","alias_value":"MU5Q3RHHUXYZ","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"pith_short_16","alias_value":"MU5Q3RHHUXYZLVLL","created_at":"2026-05-20T00:01:12Z"},{"alias_kind":"pith_short_8","alias_value":"MU5Q3RHH","created_at":"2026-05-20T00:01:12Z"}],"graph_snapshots":[{"event_id":"sha256:c3019943fe182f2228cf26c00697b209def8687c59d55abf78e898cebe05216c","target":"graph","created_at":"2026-05-20T00:01:12Z","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":"We identify a fundamental flaw in the standard TCAV score: its reliance on a discontinuous indicator function induces non-decaying variance in critical regimes. To address this, we introduce α-TCAV, a generalized framework that replaces the indicator with a parameterized smooth function, yielding a unified probabilistic formulation that subsumes both TCAV and Multi-TCAV."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The derivations of the distributions for PatternCAV, FastCAV, and ridge-regression CAVs are analytically tractable and that the smooth function in α-TCAV can be tuned to achieve either imitation of Multi-TCAV or a calibrated Bayes-optimal measure without introducing new bias. This premise enters in the stochastic analysis section and the guidance on tuning the parameter."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"α-TCAV replaces TCAV's hard indicator with a tunable smooth function to create a unified probabilistic framework with lower variance and guidance for parameter choice or Bayes-optimal scoring."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"The standard TCAV score has non-decaying variance from its discontinuous indicator, which α-TCAV fixes by substituting a parameterized smooth function for a unified probabilistic framework."}],"snapshot_sha256":"477dbfda13a66c7a4a0e37f271115698b415d418ec56aee7cca43c0279d783a7"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T20:01:19.233329Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T19:51:29.209145Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T19:33:29.223945Z","status":"skipped","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T17:21:56.044684Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.15688/integrity.json","findings":[],"snapshot_sha256":"8b1a57cb22acc2942c59f0a29d2cdd716acfafd254b0159d827b4196ec2fdaeb","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Concept Activation Vectors (CAVs) are a fundamental tool for concept-based explainability in deep learning, yet their practical utility is limited by statistical instability. We analyze the stochastic nature of CAVs and the Testing with CAVs (TCAV) method, deriving the distributions of major CAV classes including PatternCAV, FastCAV, and ridge regression-based CAVs. We then identify a fundamental flaw in the standard TCAV score: its reliance on a discontinuous indicator function induces non-decaying variance in critical regimes. To address this, we introduce $\\alpha$-TCAV, a generalized framew","authors_text":"Alexander Jung, Ekkehard Schnoor, Jawher Said, Malik Tiomoko, Wojciech Samek","cross_cats":["cs.AI","cs.LG","math.PR"],"headline":"The standard TCAV score has non-decaying variance from its discontinuous indicator, which α-TCAV fixes by substituting a parameterized smooth function for a unified probabilistic framework.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-05-15T07:22:59Z","title":"$\\alpha$-TCAV: A Unified Framework for Testing with Concept Activation Vectors"},"references":{"count":151,"internal_anchors":4,"resolved_work":151,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Zur Elektrodynamik bewegter Körper","work_id":"f62dda56-8333-47e0-bc47-249c3fbb0cbe","year":1905},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"The Annals of Applied Probability , volume=","work_id":"4393abc6-a331-4ace-a7b4-41b21e0bf3f5","year":2018},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"The Thirteenth International Conference on Learning Representations , year=","work_id":"d465d843-18f0-4777-accf-6ce6e6dccefb","year":null},{"cited_arxiv_id":"2604.03146","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Characterization of Gaussian Universality Breakdown in High-Dimensional Empirical Risk Minimization","work_id":"b7744d6f-6170-4ccf-a706-a023dc596384","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"The \\ Companion","work_id":"de8bc6e8-e1fa-4090-a818-24103cd763f9","year":1993}],"snapshot_sha256":"9e39aa12c286e40582d729de1960e9923b443772b61bd958d9497a40972a95a5"},"source":{"id":"2605.15688","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T19:44:12.106838Z","id":"1a27c302-944f-4e44-a65a-0fcc8687b08d","model_set":{"reader":"grok-4.3"},"one_line_summary":"α-TCAV replaces TCAV's hard indicator with a tunable smooth function to create a unified probabilistic framework with lower variance and guidance for parameter choice or Bayes-optimal scoring.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"The standard TCAV score has non-decaying variance from its discontinuous indicator, which α-TCAV fixes by substituting a parameterized smooth function for a unified probabilistic framework.","strongest_claim":"We identify a fundamental flaw in the standard TCAV score: its reliance on a discontinuous indicator function induces non-decaying variance in critical regimes. To address this, we introduce α-TCAV, a generalized framework that replaces the indicator with a parameterized smooth function, yielding a unified probabilistic formulation that subsumes both TCAV and Multi-TCAV.","weakest_assumption":"The derivations of the distributions for PatternCAV, FastCAV, and ridge-regression CAVs are analytically tractable and that the smooth function in α-TCAV can be tuned to achieve either imitation of Multi-TCAV or a calibrated Bayes-optimal measure without introducing new bias. This premise enters in the stochastic analysis section and the guidance on tuning the parameter."}},"verdict_id":"1a27c302-944f-4e44-a65a-0fcc8687b08d"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:79ee46ac601a2627307856bed24619bf5ba96e2bf0e88cf082d0e12ced7e07cb","target":"record","created_at":"2026-05-20T00:01:12Z","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":"a941485d7933e4d858d2edede58b35dc959ff72cc3d1707b6cdf078f35c9902a","cross_cats_sorted":["cs.AI","cs.LG","math.PR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-05-15T07:22:59Z","title_canon_sha256":"8d586ce140c497c4ae3bba821a7b9172e1272da09adfe3fcac59fdfce40bfc74"},"schema_version":"1.0","source":{"id":"2605.15688","kind":"arxiv","version":1}},"canonical_sha256":"653b0dc4e7a5f195d56be55aba87d34b128ff42f9a21de9e1c32d6a9a41f87db","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"653b0dc4e7a5f195d56be55aba87d34b128ff42f9a21de9e1c32d6a9a41f87db","first_computed_at":"2026-05-20T00:01:12.526593Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:01:12.526593Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vmEAFVJZ0NhymvIUjps8YDmgPWPVqcECemDSWb55hNOvJ5k9pqD17/vLWLG/eAuD7lD+bILuCb4Pt6phlHAdAQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:01:12.527520Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15688","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:79ee46ac601a2627307856bed24619bf5ba96e2bf0e88cf082d0e12ced7e07cb","sha256:c3019943fe182f2228cf26c00697b209def8687c59d55abf78e898cebe05216c"],"state_sha256":"978fdf84a924286a24b1055e365a7bd6b8d2ff9f491252c0aaa409ea654cad50"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RHR7NgAzx8wNCCIF+wK3rDtEtZB2TPxw7ifZGlSLTbUBvVJLX9oICP3TsoX9pe20H7ZQMOQL/Eq6QgI8HFnmDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T23:18:27.375967Z","bundle_sha256":"400454481b4f63d785e1a8b5ff211f72efed20baeece41ef491c08b3912206b2"}}