{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6X2WD2J2UBHRQVHY5XOWMJWEY4","short_pith_number":"pith:6X2WD2J2","schema_version":"1.0","canonical_sha256":"f5f561e93aa04f1854f8eddd6626c4c71d55684b7e5841588f392d9da7f4b33b","source":{"kind":"arxiv","id":"2608.13223","version":1},"attestation_state":"computed","paper":{"title":"Reliability analysis for BraTS-GoAT segmentation: a controlled robustness study of deep-ensemble uncertainty","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Le Zhang, Riya Deepak Shet","submitted_at":"2026-08-13T13:27:21Z","abstract_excerpt":"Deep networks segment brain tumours accurately in-distribution, but can fail silently when the input differs from their training data. That risk is central to clinical deployment and is the premise of the BraTS-GoAT generalizability task. We ask not only how well a model segments, but whether its uncertainty knows when it is wrong. On BraTS-GoAT (Task 3) we train a 5-fold cross-validated nnU-Net baseline (one held-out prediction per case) and a 3-seed deep ensemble. Both are evaluated for calibration and error detection on a per-region relevant mask, aggregated per case. In-distribution the 3-"},"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":"2608.13223","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-13T13:27:21Z","cross_cats_sorted":[],"title_canon_sha256":"f3427fde3436871fff8a561e092ef03cccaa45f8157b69f9160ff6e331c13f59","abstract_canon_sha256":"391074fe3a1ed22fdf78a3b545a3c0766363e6d8891ed254030b93801795c2d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-14T01:00:42.904670Z","signature_b64":"5wFUE1iARvD5pENn4Q3gzAKJEnxddaFlrEHW2eUjgoT3xaLjUG2fVmbkilUfIc1moIGVlEDd3YqjVcZpvE5rCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5f561e93aa04f1854f8eddd6626c4c71d55684b7e5841588f392d9da7f4b33b","last_reissued_at":"2026-08-14T01:00:42.902535Z","signature_status":"signed_v1","first_computed_at":"2026-08-14T01:00:42.902535Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reliability analysis for BraTS-GoAT segmentation: a controlled robustness study of deep-ensemble uncertainty","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Le Zhang, Riya Deepak Shet","submitted_at":"2026-08-13T13:27:21Z","abstract_excerpt":"Deep networks segment brain tumours accurately in-distribution, but can fail silently when the input differs from their training data. That risk is central to clinical deployment and is the premise of the BraTS-GoAT generalizability task. We ask not only how well a model segments, but whether its uncertainty knows when it is wrong. On BraTS-GoAT (Task 3) we train a 5-fold cross-validated nnU-Net baseline (one held-out prediction per case) and a 3-seed deep ensemble. Both are evaluated for calibration and error detection on a per-region relevant mask, aggregated per case. In-distribution the 3-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13223","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/2608.13223/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":"2608.13223","created_at":"2026-08-14T01:00:42.903606+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.13223v1","created_at":"2026-08-14T01:00:42.903606+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13223","created_at":"2026-08-14T01:00:42.903606+00:00"},{"alias_kind":"pith_short_12","alias_value":"6X2WD2J2UBHR","created_at":"2026-08-14T01:00:42.903606+00:00"},{"alias_kind":"pith_short_16","alias_value":"6X2WD2J2UBHRQVHY","created_at":"2026-08-14T01:00:42.903606+00:00"},{"alias_kind":"pith_short_8","alias_value":"6X2WD2J2","created_at":"2026-08-14T01:00:42.903606+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/6X2WD2J2UBHRQVHY5XOWMJWEY4","json":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4.json","graph_json":"https://pith.science/api/pith-number/6X2WD2J2UBHRQVHY5XOWMJWEY4/graph.json","events_json":"https://pith.science/api/pith-number/6X2WD2J2UBHRQVHY5XOWMJWEY4/events.json","paper":"https://pith.science/paper/6X2WD2J2"},"agent_actions":{"view_html":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4","download_json":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4.json","view_paper":"https://pith.science/paper/6X2WD2J2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.13223&json=true","fetch_graph":"https://pith.science/api/pith-number/6X2WD2J2UBHRQVHY5XOWMJWEY4/graph.json","fetch_events":"https://pith.science/api/pith-number/6X2WD2J2UBHRQVHY5XOWMJWEY4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4/action/storage_attestation","attest_author":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4/action/author_attestation","sign_citation":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4/action/citation_signature","submit_replication":"https://pith.science/pith/6X2WD2J2UBHRQVHY5XOWMJWEY4/action/replication_record"}},"created_at":"2026-08-14T01:00:42.903606+00:00","updated_at":"2026-08-14T01:00:42.903606+00:00"}