{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UOV3EJSIDX2XR36YBYKYBR2LR4","short_pith_number":"pith:UOV3EJSI","schema_version":"1.0","canonical_sha256":"a3abb226481df578efd80e1580c74b8f1eed91c40c3d8dfdbf1456d834d18fbf","source":{"kind":"arxiv","id":"2505.03205","version":3},"attestation_state":"computed","paper":{"title":"Transformers for Learning on Noisy and Task-Level Manifolds: Approximation and Generalization Insights","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA","math.ST","stat.TH"],"primary_cat":"cs.LG","authors_text":"Alexander Cloninger, Alex Havrilla, Rongjie Lai, Wenjing Liao, Zhaiming Shen","submitted_at":"2025-05-06T05:41:46Z","abstract_excerpt":"Transformers serve as the foundational architecture for large language and video generation models, such as GPT, BERT, SORA and their successors. Empirical studies have demonstrated that real-world data and learning tasks exhibit low-dimensional structures, along with some noise or measurement error. The performance of transformers tends to depend on the intrinsic dimension of the data/tasks, though theoretical understandings remain largely unexplored for transformers. This work establishes a theoretical foundation by analyzing the performance of transformers for regression tasks involving noi"},"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":"2505.03205","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-06T05:41:46Z","cross_cats_sorted":["cs.NA","math.NA","math.ST","stat.TH"],"title_canon_sha256":"66c005030f9934fd0c6e6eb499b3565dd812b2b2367703781b4b84c0cb438000","abstract_canon_sha256":"912b47d85eb3bd296e81b0bb5aa153058ac707ecf6d4a262bbbb481e3f0eba9f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:50.033785Z","signature_b64":"vcEAlLBqo84CZh2S1veY7vUyAxoEKGz8XLldMe1PAlxrjlcZblaCEyyjQIv7tRoODtU/b9beO7gKBSaghSUuAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a3abb226481df578efd80e1580c74b8f1eed91c40c3d8dfdbf1456d834d18fbf","last_reissued_at":"2026-05-20T00:02:50.033039Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:50.033039Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transformers for Learning on Noisy and Task-Level Manifolds: Approximation and Generalization Insights","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA","math.ST","stat.TH"],"primary_cat":"cs.LG","authors_text":"Alexander Cloninger, Alex Havrilla, Rongjie Lai, Wenjing Liao, Zhaiming Shen","submitted_at":"2025-05-06T05:41:46Z","abstract_excerpt":"Transformers serve as the foundational architecture for large language and video generation models, such as GPT, BERT, SORA and their successors. Empirical studies have demonstrated that real-world data and learning tasks exhibit low-dimensional structures, along with some noise or measurement error. The performance of transformers tends to depend on the intrinsic dimension of the data/tasks, though theoretical understandings remain largely unexplored for transformers. This work establishes a theoretical foundation by analyzing the performance of transformers for regression tasks involving noi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.03205","kind":"arxiv","version":3},"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/2505.03205/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":"2505.03205","created_at":"2026-05-20T00:02:50.033160+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.03205v3","created_at":"2026-05-20T00:02:50.033160+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.03205","created_at":"2026-05-20T00:02:50.033160+00:00"},{"alias_kind":"pith_short_12","alias_value":"UOV3EJSIDX2X","created_at":"2026-05-20T00:02:50.033160+00:00"},{"alias_kind":"pith_short_16","alias_value":"UOV3EJSIDX2XR36Y","created_at":"2026-05-20T00:02:50.033160+00:00"},{"alias_kind":"pith_short_8","alias_value":"UOV3EJSI","created_at":"2026-05-20T00:02:50.033160+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2510.03989","citing_title":"A Mathematical Explanation of Transformers","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05176","citing_title":"Understanding In-Context Learning for Nonlinear Regression with Transformers: Attention as Featurizer","ref_index":31,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4","json":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4.json","graph_json":"https://pith.science/api/pith-number/UOV3EJSIDX2XR36YBYKYBR2LR4/graph.json","events_json":"https://pith.science/api/pith-number/UOV3EJSIDX2XR36YBYKYBR2LR4/events.json","paper":"https://pith.science/paper/UOV3EJSI"},"agent_actions":{"view_html":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4","download_json":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4.json","view_paper":"https://pith.science/paper/UOV3EJSI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.03205&json=true","fetch_graph":"https://pith.science/api/pith-number/UOV3EJSIDX2XR36YBYKYBR2LR4/graph.json","fetch_events":"https://pith.science/api/pith-number/UOV3EJSIDX2XR36YBYKYBR2LR4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4/action/storage_attestation","attest_author":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4/action/author_attestation","sign_citation":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4/action/citation_signature","submit_replication":"https://pith.science/pith/UOV3EJSIDX2XR36YBYKYBR2LR4/action/replication_record"}},"created_at":"2026-05-20T00:02:50.033160+00:00","updated_at":"2026-05-20T00:02:50.033160+00:00"}