{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BMGSGH7WZUYK6CM7K55AF6K2DT","short_pith_number":"pith:BMGSGH7W","schema_version":"1.0","canonical_sha256":"0b0d231ff6cd30af099f577a02f95a1ce28390183a3354e39c17ab6b9e98f7dd","source":{"kind":"arxiv","id":"2602.04596","version":2},"attestation_state":"computed","paper":{"title":"Uncertainty Decomposition for Bayes-Filtered Transformers via Bayesian Predictive Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Judith Rousseau, Kenyon Ng, Sandra Fortini, Sonia Petrone, Susan Wei","submitted_at":"2026-02-04T14:23:53Z","abstract_excerpt":"Bayes-filtered transformers are transformers meta-learned on sequences from a prior predictive distribution to approximate the corresponding posterior predictive distribution. They output total predictive uncertainty in a single forward pass but never explicitly represent a posterior distribution, making the standard route to separating aleatoric from epistemic uncertainty unavailable. We address this challenge through the lens of Bayesian predictive inference (BPI). Our main result is a predictive Central Limit Theorem (CLT) for supervised settings under conditions that are among the weakest "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2602.04596","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-02-04T14:23:53Z","cross_cats_sorted":["cs.LG","stat.ME"],"title_canon_sha256":"4e14cf10fb0621768c96395633966e94df347af5a5f4e1e6a7ec08ba5816f5bd","abstract_canon_sha256":"74d5cbc6de863ee32579a79f653eb099939decef4971ce6751df715ee1db0534"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:20:43.469546Z","signature_b64":"FrQjxI36lrS1caGVbwzzVFne3acp2E7m0hlMfIiaIDjdNCN0Uk1PMsovxdJHToQUQ+L2ViQuqIcs3Fo0cYieAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b0d231ff6cd30af099f577a02f95a1ce28390183a3354e39c17ab6b9e98f7dd","last_reissued_at":"2026-07-21T01:20:43.468522Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:20:43.468522Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Uncertainty Decomposition for Bayes-Filtered Transformers via Bayesian Predictive Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Judith Rousseau, Kenyon Ng, Sandra Fortini, Sonia Petrone, Susan Wei","submitted_at":"2026-02-04T14:23:53Z","abstract_excerpt":"Bayes-filtered transformers are transformers meta-learned on sequences from a prior predictive distribution to approximate the corresponding posterior predictive distribution. They output total predictive uncertainty in a single forward pass but never explicitly represent a posterior distribution, making the standard route to separating aleatoric from epistemic uncertainty unavailable. We address this challenge through the lens of Bayesian predictive inference (BPI). Our main result is a predictive Central Limit Theorem (CLT) for supervised settings under conditions that are among the weakest "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.04596","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2602.04596/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2602.04596","created_at":"2026-07-21T01:20:43.468991+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.04596v2","created_at":"2026-07-21T01:20:43.468991+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.04596","created_at":"2026-07-21T01:20:43.468991+00:00"},{"alias_kind":"pith_short_12","alias_value":"BMGSGH7WZUYK","created_at":"2026-07-21T01:20:43.468991+00:00"},{"alias_kind":"pith_short_16","alias_value":"BMGSGH7WZUYK6CM7","created_at":"2026-07-21T01:20:43.468991+00:00"},{"alias_kind":"pith_short_8","alias_value":"BMGSGH7W","created_at":"2026-07-21T01:20:43.468991+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.00455","citing_title":"Concentration and Calibration in Predictive Bayesian Inference","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT","json":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT.json","graph_json":"https://pith.science/api/pith-number/BMGSGH7WZUYK6CM7K55AF6K2DT/graph.json","events_json":"https://pith.science/api/pith-number/BMGSGH7WZUYK6CM7K55AF6K2DT/events.json","paper":"https://pith.science/paper/BMGSGH7W"},"agent_actions":{"view_html":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT","download_json":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT.json","view_paper":"https://pith.science/paper/BMGSGH7W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.04596&json=true","fetch_graph":"https://pith.science/api/pith-number/BMGSGH7WZUYK6CM7K55AF6K2DT/graph.json","fetch_events":"https://pith.science/api/pith-number/BMGSGH7WZUYK6CM7K55AF6K2DT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT/action/storage_attestation","attest_author":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT/action/author_attestation","sign_citation":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT/action/citation_signature","submit_replication":"https://pith.science/pith/BMGSGH7WZUYK6CM7K55AF6K2DT/action/replication_record"}},"created_at":"2026-07-21T01:20:43.468991+00:00","updated_at":"2026-07-21T01:20:43.468991+00:00"}