{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TBCUVOS4WWFGG73LM4DNP2WBS5","short_pith_number":"pith:TBCUVOS4","schema_version":"1.0","canonical_sha256":"98454aba5cb58a637f6b6706d7eac19773bc6586d7c5231f69948f79d943fb21","source":{"kind":"arxiv","id":"2508.07556","version":2},"attestation_state":"computed","paper":{"title":"Uncertainty-Driven Reliability: Selective Prediction and Trustworthy Deployment in Modern Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CY","stat.ML"],"primary_cat":"cs.LG","authors_text":"Stephan Rabanser","submitted_at":"2025-08-11T02:33:53Z","abstract_excerpt":"Machine learning (ML) systems are increasingly deployed in high-stakes domains where reliability is paramount. This thesis investigates how uncertainty estimation can enhance the safety and trustworthiness of ML, focusing on selective prediction -- where models abstain when confidence is low.\n  We first show that a model's training trajectory contains rich uncertainty signals that can be exploited without altering its architecture or loss. By ensembling predictions from intermediate checkpoints, we propose a lightweight, post-hoc abstention method that works across tasks, avoids the cost of de"},"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":"2508.07556","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-11T02:33:53Z","cross_cats_sorted":["cs.AI","cs.CY","stat.ML"],"title_canon_sha256":"05d69815b3270109f778dd95318a2530e3977a81412e55ab958d5b6453f9525b","abstract_canon_sha256":"4e599f461d56b98825b9c8ccf88309608008867c491e9e266880620fb1eedc49"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:17.520651Z","signature_b64":"lgGDSTl4oi1/LiU3WFKc85ZQqbsTP0XNuTATX5HiqDjLSrEVVxxtgNA1z2m4utYy3qwWMs59lVpGwKGW64DxCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"98454aba5cb58a637f6b6706d7eac19773bc6586d7c5231f69948f79d943fb21","last_reissued_at":"2026-07-05T12:06:17.520097Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:17.520097Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Uncertainty-Driven Reliability: Selective Prediction and Trustworthy Deployment in Modern Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CY","stat.ML"],"primary_cat":"cs.LG","authors_text":"Stephan Rabanser","submitted_at":"2025-08-11T02:33:53Z","abstract_excerpt":"Machine learning (ML) systems are increasingly deployed in high-stakes domains where reliability is paramount. This thesis investigates how uncertainty estimation can enhance the safety and trustworthiness of ML, focusing on selective prediction -- where models abstain when confidence is low.\n  We first show that a model's training trajectory contains rich uncertainty signals that can be exploited without altering its architecture or loss. By ensembling predictions from intermediate checkpoints, we propose a lightweight, post-hoc abstention method that works across tasks, avoids the cost of de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.07556","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/2508.07556/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":"2508.07556","created_at":"2026-07-05T12:06:17.520160+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.07556v2","created_at":"2026-07-05T12:06:17.520160+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.07556","created_at":"2026-07-05T12:06:17.520160+00:00"},{"alias_kind":"pith_short_12","alias_value":"TBCUVOS4WWFG","created_at":"2026-07-05T12:06:17.520160+00:00"},{"alias_kind":"pith_short_16","alias_value":"TBCUVOS4WWFGG73L","created_at":"2026-07-05T12:06:17.520160+00:00"},{"alias_kind":"pith_short_8","alias_value":"TBCUVOS4","created_at":"2026-07-05T12:06:17.520160+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5","json":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5.json","graph_json":"https://pith.science/api/pith-number/TBCUVOS4WWFGG73LM4DNP2WBS5/graph.json","events_json":"https://pith.science/api/pith-number/TBCUVOS4WWFGG73LM4DNP2WBS5/events.json","paper":"https://pith.science/paper/TBCUVOS4"},"agent_actions":{"view_html":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5","download_json":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5.json","view_paper":"https://pith.science/paper/TBCUVOS4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.07556&json=true","fetch_graph":"https://pith.science/api/pith-number/TBCUVOS4WWFGG73LM4DNP2WBS5/graph.json","fetch_events":"https://pith.science/api/pith-number/TBCUVOS4WWFGG73LM4DNP2WBS5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5/action/storage_attestation","attest_author":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5/action/author_attestation","sign_citation":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5/action/citation_signature","submit_replication":"https://pith.science/pith/TBCUVOS4WWFGG73LM4DNP2WBS5/action/replication_record"}},"created_at":"2026-07-05T12:06:17.520160+00:00","updated_at":"2026-07-05T12:06:17.520160+00:00"}