{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CWGYKBMSFFTAS37QDUMAF4ERAY","short_pith_number":"pith:CWGYKBMS","schema_version":"1.0","canonical_sha256":"158d8505922966096ff01d1802f091063b7467dbfb35e0d9d216bd950a6fca6d","source":{"kind":"arxiv","id":"2405.16563","version":2},"attestation_state":"computed","paper":{"title":"Higher-Order Transformer Derivative Estimates for Explicit Pathwise Learning Guarantees","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.NA","cs.NE","math.NA","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anastasis Kratsios, Blanka Horvath, Raeid Saqur, Xuwei Yang, Yannick Limmer","submitted_at":"2024-05-26T13:19:32Z","abstract_excerpt":"An inherent challenge in computing fully-explicit generalization bounds for transformers involves obtaining covering number estimates for the given transformer class $T$. Crude estimates rely on a uniform upper bound on the local-Lipschitz constants of transformers in $T$, and finer estimates require an analysis of their higher-order partial derivatives. Unfortunately, these precise higher-order derivative estimates for (realistic) transformer models are not currently available in the literature as they are combinatorially delicate due to the intricate compositional structure of transformer bl"},"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":"2405.16563","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-26T13:19:32Z","cross_cats_sorted":["cs.NA","cs.NE","math.NA","math.PR","stat.ML"],"title_canon_sha256":"8058c0b2e6713f262a88aea3296868fa7f341370d5b8fcd70643c9389206990c","abstract_canon_sha256":"fb181df19344e1d698fd6390510eedb1f905582df4ff0b6efdc4830ce2bb3d69"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:10:23.488485Z","signature_b64":"IaR/owwScCxa0jkDqUsL/vPvhitpyusvQvCpVBEnh1WBJeZHRA+/p/UrGaSzcZjH164YXgvLchjzBcdGA25MBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"158d8505922966096ff01d1802f091063b7467dbfb35e0d9d216bd950a6fca6d","last_reissued_at":"2026-07-05T10:10:23.487917Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:10:23.487917Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Higher-Order Transformer Derivative Estimates for Explicit Pathwise Learning Guarantees","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.NA","cs.NE","math.NA","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anastasis Kratsios, Blanka Horvath, Raeid Saqur, Xuwei Yang, Yannick Limmer","submitted_at":"2024-05-26T13:19:32Z","abstract_excerpt":"An inherent challenge in computing fully-explicit generalization bounds for transformers involves obtaining covering number estimates for the given transformer class $T$. Crude estimates rely on a uniform upper bound on the local-Lipschitz constants of transformers in $T$, and finer estimates require an analysis of their higher-order partial derivatives. Unfortunately, these precise higher-order derivative estimates for (realistic) transformer models are not currently available in the literature as they are combinatorially delicate due to the intricate compositional structure of transformer bl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.16563","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/2405.16563/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":"2405.16563","created_at":"2026-07-05T10:10:23.488001+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.16563v2","created_at":"2026-07-05T10:10:23.488001+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.16563","created_at":"2026-07-05T10:10:23.488001+00:00"},{"alias_kind":"pith_short_12","alias_value":"CWGYKBMSFFTA","created_at":"2026-07-05T10:10:23.488001+00:00"},{"alias_kind":"pith_short_16","alias_value":"CWGYKBMSFFTAS37Q","created_at":"2026-07-05T10:10:23.488001+00:00"},{"alias_kind":"pith_short_8","alias_value":"CWGYKBMS","created_at":"2026-07-05T10:10:23.488001+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.07097","citing_title":"Every Feedforward Neural Network Definable in an o-Minimal Structure Has Finite Sample Complexity","ref_index":59,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY","json":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY.json","graph_json":"https://pith.science/api/pith-number/CWGYKBMSFFTAS37QDUMAF4ERAY/graph.json","events_json":"https://pith.science/api/pith-number/CWGYKBMSFFTAS37QDUMAF4ERAY/events.json","paper":"https://pith.science/paper/CWGYKBMS"},"agent_actions":{"view_html":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY","download_json":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY.json","view_paper":"https://pith.science/paper/CWGYKBMS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.16563&json=true","fetch_graph":"https://pith.science/api/pith-number/CWGYKBMSFFTAS37QDUMAF4ERAY/graph.json","fetch_events":"https://pith.science/api/pith-number/CWGYKBMSFFTAS37QDUMAF4ERAY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY/action/storage_attestation","attest_author":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY/action/author_attestation","sign_citation":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY/action/citation_signature","submit_replication":"https://pith.science/pith/CWGYKBMSFFTAS37QDUMAF4ERAY/action/replication_record"}},"created_at":"2026-07-05T10:10:23.488001+00:00","updated_at":"2026-07-05T10:10:23.488001+00:00"}