{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JUEE6EBR3QO53VPHMSYBBSCCOH","short_pith_number":"pith:JUEE6EBR","schema_version":"1.0","canonical_sha256":"4d084f1031dc1dddd5e764b010c84271f23cf2e73e1909ac118117cfc3e6a7a6","source":{"kind":"arxiv","id":"2506.01552","version":1},"attestation_state":"computed","paper":{"title":"To Each Metric Its Decoding: Post-Hoc Optimal Decision Rules of Probabilistic Hierarchical Classifiers","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexandre Perez-Lebel, Antoine Saillenfest, Matthieu Labeau, Roman Plaud, Thomas Bonald","submitted_at":"2025-06-02T11:29:40Z","abstract_excerpt":"Hierarchical classification offers an approach to incorporate the concept of mistake severity by leveraging a structured, labeled hierarchy. However, decoding in such settings frequently relies on heuristic decision rules, which may not align with task-specific evaluation metrics. In this work, we propose a framework for the optimal decoding of an output probability distribution with respect to a target metric. We derive optimal decision rules for increasingly complex prediction settings, providing universal algorithms when candidates are limited to the set of nodes. In the most general case o"},"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":"2506.01552","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.LG","submitted_at":"2025-06-02T11:29:40Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"ca62997b95ceb01ff4b9b8a6a95e7db1b584a43c6185ac65a17a17413f0845fa","abstract_canon_sha256":"6c4b969445049d2970189b039417103f71b2111ea8675b9b6d559b0eacf5c120"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:15.655564Z","signature_b64":"eLasfnVHKlL254cwQZ6Y1/48Mnpm7ZMNroa9aOfdIo3soSa5Jq5cma35hVh1ie1BPmmg7UMjCnzuzX4dKWdOAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d084f1031dc1dddd5e764b010c84271f23cf2e73e1909ac118117cfc3e6a7a6","last_reissued_at":"2026-07-05T11:14:15.655004Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:15.655004Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"To Each Metric Its Decoding: Post-Hoc Optimal Decision Rules of Probabilistic Hierarchical Classifiers","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexandre Perez-Lebel, Antoine Saillenfest, Matthieu Labeau, Roman Plaud, Thomas Bonald","submitted_at":"2025-06-02T11:29:40Z","abstract_excerpt":"Hierarchical classification offers an approach to incorporate the concept of mistake severity by leveraging a structured, labeled hierarchy. However, decoding in such settings frequently relies on heuristic decision rules, which may not align with task-specific evaluation metrics. In this work, we propose a framework for the optimal decoding of an output probability distribution with respect to a target metric. We derive optimal decision rules for increasingly complex prediction settings, providing universal algorithms when candidates are limited to the set of nodes. In the most general case o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01552","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/2506.01552/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":"2506.01552","created_at":"2026-07-05T11:14:15.655068+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.01552v1","created_at":"2026-07-05T11:14:15.655068+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01552","created_at":"2026-07-05T11:14:15.655068+00:00"},{"alias_kind":"pith_short_12","alias_value":"JUEE6EBR3QO5","created_at":"2026-07-05T11:14:15.655068+00:00"},{"alias_kind":"pith_short_16","alias_value":"JUEE6EBR3QO53VPH","created_at":"2026-07-05T11:14:15.655068+00:00"},{"alias_kind":"pith_short_8","alias_value":"JUEE6EBR","created_at":"2026-07-05T11:14:15.655068+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/JUEE6EBR3QO53VPHMSYBBSCCOH","json":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH.json","graph_json":"https://pith.science/api/pith-number/JUEE6EBR3QO53VPHMSYBBSCCOH/graph.json","events_json":"https://pith.science/api/pith-number/JUEE6EBR3QO53VPHMSYBBSCCOH/events.json","paper":"https://pith.science/paper/JUEE6EBR"},"agent_actions":{"view_html":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH","download_json":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH.json","view_paper":"https://pith.science/paper/JUEE6EBR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.01552&json=true","fetch_graph":"https://pith.science/api/pith-number/JUEE6EBR3QO53VPHMSYBBSCCOH/graph.json","fetch_events":"https://pith.science/api/pith-number/JUEE6EBR3QO53VPHMSYBBSCCOH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH/action/storage_attestation","attest_author":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH/action/author_attestation","sign_citation":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH/action/citation_signature","submit_replication":"https://pith.science/pith/JUEE6EBR3QO53VPHMSYBBSCCOH/action/replication_record"}},"created_at":"2026-07-05T11:14:15.655068+00:00","updated_at":"2026-07-05T11:14:15.655068+00:00"}