{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YQCM2AJSOCJMEEK3UB4WNURMUO","short_pith_number":"pith:YQCM2AJS","schema_version":"1.0","canonical_sha256":"c404cd01327092c2115ba07966d22ca38079fc983886fa7c032dac016303693d","source":{"kind":"arxiv","id":"2605.28420","version":1},"attestation_state":"computed","paper":{"title":"Conveyance: A Versatile Framework for Learning in Structured Class Spaces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Gr\\'egoire Montavon, Nils K\\\"orber, Yasser Taha","submitted_at":"2026-05-27T12:51:59Z","abstract_excerpt":"While machine learning (ML) architectures have evolved rapidly to account for complex data, loss functions like cross-entropy remain mostly structure-agnostic in many real-world applications. However, the `class-symmetric' nature of these standard losses fundamentally limits the ability of ML models to exploit structural relationships between classes, particularly when facing structured noise. We propose \\textsc{Conveyance}, a new classification approach and associated loss function tailored to structured class spaces. It allows users to encode graph-like relations between classes without havi"},"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":"2605.28420","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-27T12:51:59Z","cross_cats_sorted":[],"title_canon_sha256":"5003292f9ef37f70f73893b48e037a41b6beddad6d236642a70410378e89f32a","abstract_canon_sha256":"f78adbdf4df6fcbeef622023bad5d2c02d7735a43425760d75e4464ff91e2713"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:05:17.668806Z","signature_b64":"ZKj9AQHY/meo9BF4dfKICD/YsCsXJUSd4d3hWV29ICKn77qI2+6HYwBFmPdAbxGkxQrHgZL+g7S/kUZ4AsIUAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c404cd01327092c2115ba07966d22ca38079fc983886fa7c032dac016303693d","last_reissued_at":"2026-05-28T01:05:17.668398Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:05:17.668398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Conveyance: A Versatile Framework for Learning in Structured Class Spaces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Gr\\'egoire Montavon, Nils K\\\"orber, Yasser Taha","submitted_at":"2026-05-27T12:51:59Z","abstract_excerpt":"While machine learning (ML) architectures have evolved rapidly to account for complex data, loss functions like cross-entropy remain mostly structure-agnostic in many real-world applications. However, the `class-symmetric' nature of these standard losses fundamentally limits the ability of ML models to exploit structural relationships between classes, particularly when facing structured noise. We propose \\textsc{Conveyance}, a new classification approach and associated loss function tailored to structured class spaces. It allows users to encode graph-like relations between classes without havi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28420","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/2605.28420/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":"2605.28420","created_at":"2026-05-28T01:05:17.668458+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.28420v1","created_at":"2026-05-28T01:05:17.668458+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28420","created_at":"2026-05-28T01:05:17.668458+00:00"},{"alias_kind":"pith_short_12","alias_value":"YQCM2AJSOCJM","created_at":"2026-05-28T01:05:17.668458+00:00"},{"alias_kind":"pith_short_16","alias_value":"YQCM2AJSOCJMEEK3","created_at":"2026-05-28T01:05:17.668458+00:00"},{"alias_kind":"pith_short_8","alias_value":"YQCM2AJS","created_at":"2026-05-28T01:05:17.668458+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/YQCM2AJSOCJMEEK3UB4WNURMUO","json":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO.json","graph_json":"https://pith.science/api/pith-number/YQCM2AJSOCJMEEK3UB4WNURMUO/graph.json","events_json":"https://pith.science/api/pith-number/YQCM2AJSOCJMEEK3UB4WNURMUO/events.json","paper":"https://pith.science/paper/YQCM2AJS"},"agent_actions":{"view_html":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO","download_json":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO.json","view_paper":"https://pith.science/paper/YQCM2AJS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.28420&json=true","fetch_graph":"https://pith.science/api/pith-number/YQCM2AJSOCJMEEK3UB4WNURMUO/graph.json","fetch_events":"https://pith.science/api/pith-number/YQCM2AJSOCJMEEK3UB4WNURMUO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO/action/storage_attestation","attest_author":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO/action/author_attestation","sign_citation":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO/action/citation_signature","submit_replication":"https://pith.science/pith/YQCM2AJSOCJMEEK3UB4WNURMUO/action/replication_record"}},"created_at":"2026-05-28T01:05:17.668458+00:00","updated_at":"2026-05-28T01:05:17.668458+00:00"}