{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:7JVW7MQ7KD7ONX7VG6APWPLAXY","short_pith_number":"pith:7JVW7MQ7","canonical_record":{"source":{"id":"2605.16645","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.ST","submitted_at":"2026-05-15T21:33:38Z","cross_cats_sorted":["cs.IT","cs.LG","math.IT","stat.ML","stat.TH"],"title_canon_sha256":"0bac3f9c2c2fabee2359b15caeb17b03ee20ee175c4999bb8075d9ce849a1708","abstract_canon_sha256":"f8f4f8d837ff4722cdd7fe6dae8a1c16ac4b9e5fd4886e3856dad02d542eb934"},"schema_version":"1.0"},"canonical_sha256":"fa6b6fb21f50fee6dff53780fb3d60be07392612b84d96c983fb42366df961c1","source":{"kind":"arxiv","id":"2605.16645","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16645","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16645v1","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16645","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"pith_short_12","alias_value":"7JVW7MQ7KD7O","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"pith_short_16","alias_value":"7JVW7MQ7KD7ONX7V","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"pith_short_8","alias_value":"7JVW7MQ7","created_at":"2026-05-20T00:02:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:7JVW7MQ7KD7ONX7VG6APWPLAXY","target":"record","payload":{"canonical_record":{"source":{"id":"2605.16645","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.ST","submitted_at":"2026-05-15T21:33:38Z","cross_cats_sorted":["cs.IT","cs.LG","math.IT","stat.ML","stat.TH"],"title_canon_sha256":"0bac3f9c2c2fabee2359b15caeb17b03ee20ee175c4999bb8075d9ce849a1708","abstract_canon_sha256":"f8f4f8d837ff4722cdd7fe6dae8a1c16ac4b9e5fd4886e3856dad02d542eb934"},"schema_version":"1.0"},"canonical_sha256":"fa6b6fb21f50fee6dff53780fb3d60be07392612b84d96c983fb42366df961c1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:34.135923Z","signature_b64":"H2FsFOZOpsZ7a6VWMli+HgEjvNN1M2lVBWVLuO95QdkIGkaMEQ8hTol8faaz9HeuiDlYjm+Yd2zRX4Wq0AqRCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa6b6fb21f50fee6dff53780fb3d60be07392612b84d96c983fb42366df961c1","last_reissued_at":"2026-05-20T00:02:34.135065Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:34.135065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.16645","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":"2026-05-20T00:02:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tasq/gEhYJx2EakrcB+8JjFxd5g6CBAnoOpJckkCIeJ2NULhkz6LNYXbZ/I87ZfTMNkhDCQMFpnThPJG3TInCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T20:44:49.509754Z"},"content_sha256":"bcee9a33fcf395eba5c1cacaa7d0962018c178b1970f7c22e2660f1138c60c9f","schema_version":"1.0","event_id":"sha256:bcee9a33fcf395eba5c1cacaa7d0962018c178b1970f7c22e2660f1138c60c9f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:7JVW7MQ7KD7ONX7VG6APWPLAXY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Statistical Unlearning of Distributions: A Hypothesis Testing Approach","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A hypothesis test comparing edited data to desired and unwanted distributions supplies a criterion for choosing which samples to remove when unlearning entire domains.","cross_cats":["cs.IT","cs.LG","math.IT","stat.ML","stat.TH"],"primary_cat":"math.ST","authors_text":"Aaradhya Pandey, Sanjeev Kulkarni","submitted_at":"2026-05-15T21:33:38Z","abstract_excerpt":"Machine learning systems increasingly face requirements to forget not only individual data points, but entire domains of information, such as toxic language, copyrighted corpora, or demographic biases. This raises a fundamental dilemma of statistical-computational tradeoffs: removing all samples from an unwanted domain may be computationally prohibitive, while randomly removing a subset may not provide distribution-level statistical guarantees. We propose a statistical framework for distributional unlearning, in which domains are modeled as probability distributions, and the goal is to remove "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We formalize this using a hypothesis test of the edited data with the desired and unwanted domains, leading to an interpretable and robust criterion for selecting samples to remove. Within this statistical framework, we characterize the fundamental region of the allowable edited data distributions and the removal-preservation Pareto frontier for a broad class of distribution families.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"Domains of information can be accurately modeled as probability distributions, and a hypothesis test between the edited dataset and the desired versus unwanted distributions provides a sufficient criterion for sample removal that preserves performance on the target domain.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A hypothesis testing approach to distributional unlearning that characterizes allowable edited distributions and removal-preservation Pareto frontiers for parametric and nonparametric families including Gaussians, Poisson, and Gaussian white noise.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A hypothesis test comparing edited data to desired and unwanted distributions supplies a criterion for choosing which samples to remove when unlearning entire domains.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e0afb824f1dc4a1ee7a70e9675b5eaf0ca8991d191b2bf3468833366ae878111"},"source":{"id":"2605.16645","kind":"arxiv","version":1},"verdict":{"id":"dd52c8dc-f6e8-4379-8779-0208adf6e342","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T20:33:57.291051Z","strongest_claim":"We formalize this using a hypothesis test of the edited data with the desired and unwanted domains, leading to an interpretable and robust criterion for selecting samples to remove. Within this statistical framework, we characterize the fundamental region of the allowable edited data distributions and the removal-preservation Pareto frontier for a broad class of distribution families.","one_line_summary":"A hypothesis testing approach to distributional unlearning that characterizes allowable edited distributions and removal-preservation Pareto frontiers for parametric and nonparametric families including Gaussians, Poisson, and Gaussian white noise.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"Domains of information can be accurately modeled as probability distributions, and a hypothesis test between the edited dataset and the desired versus unwanted distributions provides a sufficient criterion for sample removal that preserves performance on the target domain.","pith_extraction_headline":"A hypothesis test comparing edited data to desired and unwanted distributions supplies a criterion for choosing which samples to remove when unlearning entire domains."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16645/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T21:01:19.317906Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T20:41:57.812891Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.409938Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.528905Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"eedc26da4df8d8f3f30b616cfb06ab0d5e039fc8931c27aebee42672c8488248"},"references":{"count":13,"sample":[{"doi":"","year":2024,"title":"Sangamesh Kodge, Gobinda Saha, and Kaushik Roy","work_id":"4c1a7271-0da3-4790-9edb-c1e84fb0ac31","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/978-3-031-26979-0_15","year":2023,"title":"Vladimir Koltchinskii and Martin Wahl","work_id":"94e68095-b181-4818-8476-da80221120f6","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"T(t 1 +X, t 2 +X)(x) =F F −1(1−x)−δ =T(t 2 +X, t 1 +X)(x).(25) So,T(t 1 +X, t 2 +X)(x) =T(X, δ+X)(x) =T(δ+X, X)(x)(26)","work_id":"3b6bda73-9c2a-4277-bba0-655813173dc6","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Thenε 1 ≤ε 2 if and only if fX,ε1(x)≥f X,ε2(x)for allx∈[0,1].(27) Intuition and importance:The proof of this lemma is along the same lines of the proof given in Dong et al","work_id":"ed90cd4a-edfa-42e5-b6b6-94ba4003fc64","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"There exists a Markov kernel 3 R : (W,F W)→(W ′,F W ′)such that R◦P=P ′ andR◦Q=Q ′,where for a probability distribution p on (W,F W), we define the probability distribution R◦p as R◦p(A ′) := Z W R(w,","work_id":"c990bba1-46a6-43b4-b96a-48124cb63203","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":13,"snapshot_sha256":"c05b95cc57f890f3aa70c60323aee418a6b9608e9e285e2116d534e00501e2ed","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a7b714ae43d69c818b99c7d95297e91f93f24dc007857eece9f2e2704d02e977"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"dd52c8dc-f6e8-4379-8779-0208adf6e342"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:02:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"52MBoJjmoh8/1Ri9zCMjrnmwHlASIn2+gldyrGa9ZxvbsZP1iwVdhFxoYwiF9YB95OCeo2MI4x02k7YnmmbCAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T20:44:49.510644Z"},"content_sha256":"ad84fc58a7726915399a31b840842ed408adc996f95a156ca04ba3eb120a00ff","schema_version":"1.0","event_id":"sha256:ad84fc58a7726915399a31b840842ed408adc996f95a156ca04ba3eb120a00ff"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7JVW7MQ7KD7ONX7VG6APWPLAXY/bundle.json","state_url":"https://pith.science/pith/7JVW7MQ7KD7ONX7VG6APWPLAXY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7JVW7MQ7KD7ONX7VG6APWPLAXY/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":"2026-05-27T20:44:49Z","links":{"resolver":"https://pith.science/pith/7JVW7MQ7KD7ONX7VG6APWPLAXY","bundle":"https://pith.science/pith/7JVW7MQ7KD7ONX7VG6APWPLAXY/bundle.json","state":"https://pith.science/pith/7JVW7MQ7KD7ONX7VG6APWPLAXY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7JVW7MQ7KD7ONX7VG6APWPLAXY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7JVW7MQ7KD7ONX7VG6APWPLAXY","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":"f8f4f8d837ff4722cdd7fe6dae8a1c16ac4b9e5fd4886e3856dad02d542eb934","cross_cats_sorted":["cs.IT","cs.LG","math.IT","stat.ML","stat.TH"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.ST","submitted_at":"2026-05-15T21:33:38Z","title_canon_sha256":"0bac3f9c2c2fabee2359b15caeb17b03ee20ee175c4999bb8075d9ce849a1708"},"schema_version":"1.0","source":{"id":"2605.16645","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16645","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16645v1","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16645","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"pith_short_12","alias_value":"7JVW7MQ7KD7O","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"pith_short_16","alias_value":"7JVW7MQ7KD7ONX7V","created_at":"2026-05-20T00:02:34Z"},{"alias_kind":"pith_short_8","alias_value":"7JVW7MQ7","created_at":"2026-05-20T00:02:34Z"}],"graph_snapshots":[{"event_id":"sha256:ad84fc58a7726915399a31b840842ed408adc996f95a156ca04ba3eb120a00ff","target":"graph","created_at":"2026-05-20T00:02:34Z","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 formalize this using a hypothesis test of the edited data with the desired and unwanted domains, leading to an interpretable and robust criterion for selecting samples to remove. Within this statistical framework, we characterize the fundamental region of the allowable edited data distributions and the removal-preservation Pareto frontier for a broad class of distribution families."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"Domains of information can be accurately modeled as probability distributions, and a hypothesis test between the edited dataset and the desired versus unwanted distributions provides a sufficient criterion for sample removal that preserves performance on the target domain."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A hypothesis testing approach to distributional unlearning that characterizes allowable edited distributions and removal-preservation Pareto frontiers for parametric and nonparametric families including Gaussians, Poisson, and Gaussian white noise."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A hypothesis test comparing edited data to desired and unwanted distributions supplies a criterion for choosing which samples to remove when unlearning entire domains."}],"snapshot_sha256":"e0afb824f1dc4a1ee7a70e9675b5eaf0ca8991d191b2bf3468833366ae878111"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"a7b714ae43d69c818b99c7d95297e91f93f24dc007857eece9f2e2704d02e977"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T21:01:19.317906Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T20:41:57.812891Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.409938Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.528905Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.16645/integrity.json","findings":[],"snapshot_sha256":"eedc26da4df8d8f3f30b616cfb06ab0d5e039fc8931c27aebee42672c8488248","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Machine learning systems increasingly face requirements to forget not only individual data points, but entire domains of information, such as toxic language, copyrighted corpora, or demographic biases. This raises a fundamental dilemma of statistical-computational tradeoffs: removing all samples from an unwanted domain may be computationally prohibitive, while randomly removing a subset may not provide distribution-level statistical guarantees. We propose a statistical framework for distributional unlearning, in which domains are modeled as probability distributions, and the goal is to remove ","authors_text":"Aaradhya Pandey, Sanjeev Kulkarni","cross_cats":["cs.IT","cs.LG","math.IT","stat.ML","stat.TH"],"headline":"A hypothesis test comparing edited data to desired and unwanted distributions supplies a criterion for choosing which samples to remove when unlearning entire domains.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.ST","submitted_at":"2026-05-15T21:33:38Z","title":"Statistical Unlearning of Distributions: A Hypothesis Testing Approach"},"references":{"count":13,"internal_anchors":0,"resolved_work":13,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Sangamesh Kodge, Gobinda Saha, and Kaushik Roy","work_id":"4c1a7271-0da3-4790-9edb-c1e84fb0ac31","year":2024},{"cited_arxiv_id":"","doi":"10.1007/978-3-031-26979-0_15","is_internal_anchor":false,"ref_index":2,"title":"Vladimir Koltchinskii and Martin Wahl","work_id":"94e68095-b181-4818-8476-da80221120f6","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"T(t 1 +X, t 2 +X)(x) =F F −1(1−x)−δ =T(t 2 +X, t 1 +X)(x).(25) So,T(t 1 +X, t 2 +X)(x) =T(X, δ+X)(x) =T(δ+X, X)(x)(26)","work_id":"3b6bda73-9c2a-4277-bba0-655813173dc6","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Thenε 1 ≤ε 2 if and only if fX,ε1(x)≥f X,ε2(x)for allx∈[0,1].(27) Intuition and importance:The proof of this lemma is along the same lines of the proof given in Dong et al","work_id":"ed90cd4a-edfa-42e5-b6b6-94ba4003fc64","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"There exists a Markov kernel 3 R : (W,F W)→(W ′,F W ′)such that R◦P=P ′ andR◦Q=Q ′,where for a probability distribution p on (W,F W), we define the probability distribution R◦p as R◦p(A ′) := Z W R(w,","work_id":"c990bba1-46a6-43b4-b96a-48124cb63203","year":2022}],"snapshot_sha256":"c05b95cc57f890f3aa70c60323aee418a6b9608e9e285e2116d534e00501e2ed"},"source":{"id":"2605.16645","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T20:33:57.291051Z","id":"dd52c8dc-f6e8-4379-8779-0208adf6e342","model_set":{"reader":"grok-4.3"},"one_line_summary":"A hypothesis testing approach to distributional unlearning that characterizes allowable edited distributions and removal-preservation Pareto frontiers for parametric and nonparametric families including Gaussians, Poisson, and Gaussian white noise.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A hypothesis test comparing edited data to desired and unwanted distributions supplies a criterion for choosing which samples to remove when unlearning entire domains.","strongest_claim":"We formalize this using a hypothesis test of the edited data with the desired and unwanted domains, leading to an interpretable and robust criterion for selecting samples to remove. Within this statistical framework, we characterize the fundamental region of the allowable edited data distributions and the removal-preservation Pareto frontier for a broad class of distribution families.","weakest_assumption":"Domains of information can be accurately modeled as probability distributions, and a hypothesis test between the edited dataset and the desired versus unwanted distributions provides a sufficient criterion for sample removal that preserves performance on the target domain."}},"verdict_id":"dd52c8dc-f6e8-4379-8779-0208adf6e342"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:bcee9a33fcf395eba5c1cacaa7d0962018c178b1970f7c22e2660f1138c60c9f","target":"record","created_at":"2026-05-20T00:02:34Z","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":"f8f4f8d837ff4722cdd7fe6dae8a1c16ac4b9e5fd4886e3856dad02d542eb934","cross_cats_sorted":["cs.IT","cs.LG","math.IT","stat.ML","stat.TH"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"math.ST","submitted_at":"2026-05-15T21:33:38Z","title_canon_sha256":"0bac3f9c2c2fabee2359b15caeb17b03ee20ee175c4999bb8075d9ce849a1708"},"schema_version":"1.0","source":{"id":"2605.16645","kind":"arxiv","version":1}},"canonical_sha256":"fa6b6fb21f50fee6dff53780fb3d60be07392612b84d96c983fb42366df961c1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fa6b6fb21f50fee6dff53780fb3d60be07392612b84d96c983fb42366df961c1","first_computed_at":"2026-05-20T00:02:34.135065Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:02:34.135065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"H2FsFOZOpsZ7a6VWMli+HgEjvNN1M2lVBWVLuO95QdkIGkaMEQ8hTol8faaz9HeuiDlYjm+Yd2zRX4Wq0AqRCw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:02:34.135923Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16645","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bcee9a33fcf395eba5c1cacaa7d0962018c178b1970f7c22e2660f1138c60c9f","sha256:ad84fc58a7726915399a31b840842ed408adc996f95a156ca04ba3eb120a00ff"],"state_sha256":"ed3a8ff38a754ae067dbb2910441de301b17861fa4612f3200adaccb8659577e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PLLEKUU9n6czsOCVrvtBkMO8CB0qL4RYumEJd/yyZNxrnf+gY/KzaGfIeVm0kJHp/E9a6AIIH4BX9+cCTjpqBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T20:44:49.514327Z","bundle_sha256":"58069e4482c5e6a3adab3762f89aa9742e080f3c19383de9089916ea1b97add1"}}