{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YI3F63DRGPPIVL7DMRFDYBBICZ","short_pith_number":"pith:YI3F63DR","schema_version":"1.0","canonical_sha256":"c2365f6c7133de8aafe3644a3c0428167c8e6687f24f2b3cc6df27cef8830702","source":{"kind":"arxiv","id":"2605.17795","version":1},"attestation_state":"computed","paper":{"title":"When Accuracy Is Not Enough: Uncertainty Collapse between Noisy Label Learning and Out-of-Distribution Detection","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Jingyang Mao, Ningkang Peng, Peirong Ma, Runhan Zhou, Yanhui Gu","submitted_at":"2026-05-18T03:19:45Z","abstract_excerpt":"Learning with noisy labels (LNL) is typically benchmarked by closed-set classification accuracy, yet deployment often requires classifiers to reject out-of-distribution (OOD) inputs. We present a learner-agnostic ACC-OOD benchmark that freezes LNL checkpoints and evaluates them with standardized near-/far-OOD routing and post-hoc scores across synthetic and real label noise. The benchmark reveals a recurring failure mode: high closed-set accuracy does not ensure OOD reliability, because low-confidence, misclassified in-distribution samples can overlap the score and feature regions occupied by "},"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":"2605.17795","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T03:19:45Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"192f82556bb93515f4154e6330fbe88f7512a06bb7d7642018988b33240fc945","abstract_canon_sha256":"5f5abf0ae47a787bbc15898d0d075fb86d44247d28fee8bd640cb931aa796a5a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:58.711970Z","signature_b64":"5/eOgmu/Z02Q/pqLi4l7E6Tg94QvLn5P0Fom1g4dCvD5EypsUOY9qMjh/6pOLxQewnshz9vLtlRqMgrQmzTdCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2365f6c7133de8aafe3644a3c0428167c8e6687f24f2b3cc6df27cef8830702","last_reissued_at":"2026-05-20T00:04:58.711044Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:58.711044Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Accuracy Is Not Enough: Uncertainty Collapse between Noisy Label Learning and Out-of-Distribution Detection","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Jingyang Mao, Ningkang Peng, Peirong Ma, Runhan Zhou, Yanhui Gu","submitted_at":"2026-05-18T03:19:45Z","abstract_excerpt":"Learning with noisy labels (LNL) is typically benchmarked by closed-set classification accuracy, yet deployment often requires classifiers to reject out-of-distribution (OOD) inputs. We present a learner-agnostic ACC-OOD benchmark that freezes LNL checkpoints and evaluates them with standardized near-/far-OOD routing and post-hoc scores across synthetic and real label noise. The benchmark reveals a recurring failure mode: high closed-set accuracy does not ensure OOD reliability, because low-confidence, misclassified in-distribution samples can overlap the score and feature regions occupied by "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17795","kind":"arxiv","version":1},"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/2605.17795/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":"2605.17795","created_at":"2026-05-20T00:04:58.711194+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.17795v1","created_at":"2026-05-20T00:04:58.711194+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17795","created_at":"2026-05-20T00:04:58.711194+00:00"},{"alias_kind":"pith_short_12","alias_value":"YI3F63DRGPPI","created_at":"2026-05-20T00:04:58.711194+00:00"},{"alias_kind":"pith_short_16","alias_value":"YI3F63DRGPPIVL7D","created_at":"2026-05-20T00:04:58.711194+00:00"},{"alias_kind":"pith_short_8","alias_value":"YI3F63DR","created_at":"2026-05-20T00:04:58.711194+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/YI3F63DRGPPIVL7DMRFDYBBICZ","json":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ.json","graph_json":"https://pith.science/api/pith-number/YI3F63DRGPPIVL7DMRFDYBBICZ/graph.json","events_json":"https://pith.science/api/pith-number/YI3F63DRGPPIVL7DMRFDYBBICZ/events.json","paper":"https://pith.science/paper/YI3F63DR"},"agent_actions":{"view_html":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ","download_json":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ.json","view_paper":"https://pith.science/paper/YI3F63DR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.17795&json=true","fetch_graph":"https://pith.science/api/pith-number/YI3F63DRGPPIVL7DMRFDYBBICZ/graph.json","fetch_events":"https://pith.science/api/pith-number/YI3F63DRGPPIVL7DMRFDYBBICZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ/action/storage_attestation","attest_author":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ/action/author_attestation","sign_citation":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ/action/citation_signature","submit_replication":"https://pith.science/pith/YI3F63DRGPPIVL7DMRFDYBBICZ/action/replication_record"}},"created_at":"2026-05-20T00:04:58.711194+00:00","updated_at":"2026-05-20T00:04:58.711194+00:00"}