{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WRXC4KWDXGDPQCE6I6G4SXL3SS","short_pith_number":"pith:WRXC4KWD","schema_version":"1.0","canonical_sha256":"b46e2e2ac3b986f8089e478dc95d7b9499542362a118764d0fe9d4e0aef9654c","source":{"kind":"arxiv","id":"2305.15508","version":4},"attestation_state":"computed","paper":{"title":"How to Fix a Broken Confidence Estimator: Evaluating Post-hoc Methods for Selective Classification with Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Danilo Silva, Lu\\'is Felipe P. Cattelan","submitted_at":"2023-05-24T18:56:55Z","abstract_excerpt":"This paper addresses the problem of selective classification for deep neural networks, where a model is allowed to abstain from low-confidence predictions to avoid potential errors. We focus on so-called post-hoc methods, which replace the confidence estimator of a given classifier without modifying or retraining it, thus being practically appealing. Considering neural networks with softmax outputs, our goal is to identify the best confidence estimator that can be computed directly from the unnormalized logits. This problem is motivated by the intriguing observation in recent work that many cl"},"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":"2305.15508","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-05-24T18:56:55Z","cross_cats_sorted":[],"title_canon_sha256":"9a7af19c15afebd70bc54b0b21a9d654ef933f1d5c004e94c4ae87aa0a0f1343","abstract_canon_sha256":"1fdfdbab7c2f13e3510bbf0c3aab5ba05713324a09430f50d0676d5be4c9bd74"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:00.568489Z","signature_b64":"rwf2vHiKrHztdm08x8zbAywof94rip1fIzULoSaJ5ZKx7x/iIuK88RCkqdmek6lfexF8CAU2ru/+RDswzfO+AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b46e2e2ac3b986f8089e478dc95d7b9499542362a118764d0fe9d4e0aef9654c","last_reissued_at":"2026-07-05T11:24:00.567960Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:00.567960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How to Fix a Broken Confidence Estimator: Evaluating Post-hoc Methods for Selective Classification with Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Danilo Silva, Lu\\'is Felipe P. Cattelan","submitted_at":"2023-05-24T18:56:55Z","abstract_excerpt":"This paper addresses the problem of selective classification for deep neural networks, where a model is allowed to abstain from low-confidence predictions to avoid potential errors. We focus on so-called post-hoc methods, which replace the confidence estimator of a given classifier without modifying or retraining it, thus being practically appealing. Considering neural networks with softmax outputs, our goal is to identify the best confidence estimator that can be computed directly from the unnormalized logits. This problem is motivated by the intriguing observation in recent work that many cl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.15508","kind":"arxiv","version":4},"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/2305.15508/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":"2305.15508","created_at":"2026-07-05T11:24:00.568024+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.15508v4","created_at":"2026-07-05T11:24:00.568024+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.15508","created_at":"2026-07-05T11:24:00.568024+00:00"},{"alias_kind":"pith_short_12","alias_value":"WRXC4KWDXGDP","created_at":"2026-07-05T11:24:00.568024+00:00"},{"alias_kind":"pith_short_16","alias_value":"WRXC4KWDXGDPQCE6","created_at":"2026-07-05T11:24:00.568024+00:00"},{"alias_kind":"pith_short_8","alias_value":"WRXC4KWD","created_at":"2026-07-05T11:24:00.568024+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02876","citing_title":"RRISE: Robust Radius Inference via a Surrogate Estimator","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19133","citing_title":"Knowing When Not to Predict: Self Supervised Learning and Abstention for Safer DR Screening","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS","json":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS.json","graph_json":"https://pith.science/api/pith-number/WRXC4KWDXGDPQCE6I6G4SXL3SS/graph.json","events_json":"https://pith.science/api/pith-number/WRXC4KWDXGDPQCE6I6G4SXL3SS/events.json","paper":"https://pith.science/paper/WRXC4KWD"},"agent_actions":{"view_html":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS","download_json":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS.json","view_paper":"https://pith.science/paper/WRXC4KWD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.15508&json=true","fetch_graph":"https://pith.science/api/pith-number/WRXC4KWDXGDPQCE6I6G4SXL3SS/graph.json","fetch_events":"https://pith.science/api/pith-number/WRXC4KWDXGDPQCE6I6G4SXL3SS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS/action/storage_attestation","attest_author":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS/action/author_attestation","sign_citation":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS/action/citation_signature","submit_replication":"https://pith.science/pith/WRXC4KWDXGDPQCE6I6G4SXL3SS/action/replication_record"}},"created_at":"2026-07-05T11:24:00.568024+00:00","updated_at":"2026-07-05T11:24:00.568024+00:00"}