{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:G74ZR2ENSVDKYQCPBIRM2O57SA","short_pith_number":"pith:G74ZR2EN","schema_version":"1.0","canonical_sha256":"37f998e88d9546ac404f0a22cd3bbf900a1f9feae586a1501a5896333e82e845","source":{"kind":"arxiv","id":"2410.24029","version":1},"attestation_state":"computed","paper":{"title":"Joint Training for Selective Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Rebecca J. Passonneau, Zhaohui Li","submitted_at":"2024-10-31T15:28:26Z","abstract_excerpt":"Classifier models are prevalent in natural language processing (NLP), often with high accuracy. Yet in real world settings, human-in-the-loop systems can foster trust in model outputs and even higher performance. Selective Prediction (SP) methods determine when to adopt a classifier's output versus defer to a human. Previous SP approaches have addressed how to improve softmax as a measure of model confidence, or have developed separate confidence estimators. One previous method involves learning a deferral model based on engineered features. We introduce a novel joint-training approach that si"},"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":"2410.24029","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-31T15:28:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"abd1a094658eeb9acca193fec877350ca3f9c86be122ff942ced57e0e7acdbfa","abstract_canon_sha256":"9bead8f61efd9fef58b15ebc96a69c3d3774a24fd9233c39a3a728c8b47cef95"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:29:18.208631Z","signature_b64":"qkOq0KlnRrZmW1DSxYigvzQgycW6Bp0So2R9P4UaPiRP+1s7jSoqjGEhuGzgft7EaH3xC/caYcfK49ocZqLeAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"37f998e88d9546ac404f0a22cd3bbf900a1f9feae586a1501a5896333e82e845","last_reissued_at":"2026-07-05T09:29:18.208091Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:29:18.208091Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Joint Training for Selective Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Rebecca J. Passonneau, Zhaohui Li","submitted_at":"2024-10-31T15:28:26Z","abstract_excerpt":"Classifier models are prevalent in natural language processing (NLP), often with high accuracy. Yet in real world settings, human-in-the-loop systems can foster trust in model outputs and even higher performance. Selective Prediction (SP) methods determine when to adopt a classifier's output versus defer to a human. Previous SP approaches have addressed how to improve softmax as a measure of model confidence, or have developed separate confidence estimators. One previous method involves learning a deferral model based on engineered features. We introduce a novel joint-training approach that si"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.24029","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/2410.24029/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":"2410.24029","created_at":"2026-07-05T09:29:18.208151+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.24029v1","created_at":"2026-07-05T09:29:18.208151+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.24029","created_at":"2026-07-05T09:29:18.208151+00:00"},{"alias_kind":"pith_short_12","alias_value":"G74ZR2ENSVDK","created_at":"2026-07-05T09:29:18.208151+00:00"},{"alias_kind":"pith_short_16","alias_value":"G74ZR2ENSVDKYQCP","created_at":"2026-07-05T09:29:18.208151+00:00"},{"alias_kind":"pith_short_8","alias_value":"G74ZR2EN","created_at":"2026-07-05T09:29:18.208151+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.07617","citing_title":"On the Limits of Selective AI Prediction: A Case Study in Clinical Decision Making","ref_index":33,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA","json":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA.json","graph_json":"https://pith.science/api/pith-number/G74ZR2ENSVDKYQCPBIRM2O57SA/graph.json","events_json":"https://pith.science/api/pith-number/G74ZR2ENSVDKYQCPBIRM2O57SA/events.json","paper":"https://pith.science/paper/G74ZR2EN"},"agent_actions":{"view_html":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA","download_json":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA.json","view_paper":"https://pith.science/paper/G74ZR2EN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.24029&json=true","fetch_graph":"https://pith.science/api/pith-number/G74ZR2ENSVDKYQCPBIRM2O57SA/graph.json","fetch_events":"https://pith.science/api/pith-number/G74ZR2ENSVDKYQCPBIRM2O57SA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA/action/storage_attestation","attest_author":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA/action/author_attestation","sign_citation":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA/action/citation_signature","submit_replication":"https://pith.science/pith/G74ZR2ENSVDKYQCPBIRM2O57SA/action/replication_record"}},"created_at":"2026-07-05T09:29:18.208151+00:00","updated_at":"2026-07-05T09:29:18.208151+00:00"}