{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZFK4HRXIQWNNEQRT4KJUX37YYI","short_pith_number":"pith:ZFK4HRXI","schema_version":"1.0","canonical_sha256":"c955c3c6e8859ad24233e2934beff8c20e276102b221756ae61b31f98d32a829","source":{"kind":"arxiv","id":"2407.07821","version":2},"attestation_state":"computed","paper":{"title":"When to Accept Automated Predictions and When to Defer to Human Judgment?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Artur Garcez, Daniel Sikar, Kaleem Peeroo, Robin Bloomfield, Tillman Weyde","submitted_at":"2024-07-10T16:45:52Z","abstract_excerpt":"Ensuring the reliability and safety of automated decision-making is crucial. It is well-known that data distribution shifts in machine learning can produce unreliable outcomes. This paper proposes a new approach for measuring the reliability of predictions under distribution shifts. We analyze how the outputs of a trained neural network change using clustering to measure distances between outputs and class centroids. We propose this distance as a metric to evaluate the confidence of predictions under distribution shifts. We assign each prediction to a cluster with centroid representing the mea"},"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":"2407.07821","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-10T16:45:52Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"54a1ba4ad41126d71e65e21bc5757c48b6ead5b2b5d4620656e405d6afb39f71","abstract_canon_sha256":"b2effaf5c231aaeb35d2e792f4d6de8e2049175bc324cee5cff1c629cf583418"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:54:53.912238Z","signature_b64":"bIqwpeBD2bdNmUyAg9jLZoQzrUlGxC0Br16hJO6c3bHhuqCc4gSroiz5jJih5t/HcVe8RjmDvUMkLkehZX8RAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c955c3c6e8859ad24233e2934beff8c20e276102b221756ae61b31f98d32a829","last_reissued_at":"2026-07-05T08:54:53.911828Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:54:53.911828Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When to Accept Automated Predictions and When to Defer to Human Judgment?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Artur Garcez, Daniel Sikar, Kaleem Peeroo, Robin Bloomfield, Tillman Weyde","submitted_at":"2024-07-10T16:45:52Z","abstract_excerpt":"Ensuring the reliability and safety of automated decision-making is crucial. It is well-known that data distribution shifts in machine learning can produce unreliable outcomes. This paper proposes a new approach for measuring the reliability of predictions under distribution shifts. We analyze how the outputs of a trained neural network change using clustering to measure distances between outputs and class centroids. We propose this distance as a metric to evaluate the confidence of predictions under distribution shifts. We assign each prediction to a cluster with centroid representing the mea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.07821","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/2407.07821/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":"2407.07821","created_at":"2026-07-05T08:54:53.911882+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.07821v2","created_at":"2026-07-05T08:54:53.911882+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.07821","created_at":"2026-07-05T08:54:53.911882+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZFK4HRXIQWNN","created_at":"2026-07-05T08:54:53.911882+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZFK4HRXIQWNNEQRT","created_at":"2026-07-05T08:54:53.911882+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZFK4HRXI","created_at":"2026-07-05T08:54:53.911882+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.21424","citing_title":"Unsupervised Incremental Learning Using Confidence-Based Pseudo-Labels","ref_index":40,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI","json":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI.json","graph_json":"https://pith.science/api/pith-number/ZFK4HRXIQWNNEQRT4KJUX37YYI/graph.json","events_json":"https://pith.science/api/pith-number/ZFK4HRXIQWNNEQRT4KJUX37YYI/events.json","paper":"https://pith.science/paper/ZFK4HRXI"},"agent_actions":{"view_html":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI","download_json":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI.json","view_paper":"https://pith.science/paper/ZFK4HRXI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.07821&json=true","fetch_graph":"https://pith.science/api/pith-number/ZFK4HRXIQWNNEQRT4KJUX37YYI/graph.json","fetch_events":"https://pith.science/api/pith-number/ZFK4HRXIQWNNEQRT4KJUX37YYI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI/action/storage_attestation","attest_author":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI/action/author_attestation","sign_citation":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI/action/citation_signature","submit_replication":"https://pith.science/pith/ZFK4HRXIQWNNEQRT4KJUX37YYI/action/replication_record"}},"created_at":"2026-07-05T08:54:53.911882+00:00","updated_at":"2026-07-05T08:54:53.911882+00:00"}