{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ICPMDFKKA34563LXG5Z6GSOJOV","short_pith_number":"pith:ICPMDFKK","schema_version":"1.0","canonical_sha256":"409ec1954a06f9df6d773773e349c9756994a87409637f4eedf7c2e290eb1a40","source":{"kind":"arxiv","id":"2105.14995","version":4},"attestation_state":"computed","paper":{"title":"Choose a Transformer: Fourier or Galerkin","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA"],"primary_cat":"cs.LG","authors_text":"Shuhao Cao","submitted_at":"2021-05-31T14:30:53Z","abstract_excerpt":"In this paper, we apply the self-attention from the state-of-the-art Transformer in Attention Is All You Need for the first time to a data-driven operator learning problem related to partial differential equations. An effort is put together to explain the heuristics of, and to improve the efficacy of the attention mechanism. By employing the operator approximation theory in Hilbert spaces, it is demonstrated for the first time that the softmax normalization in the scaled dot-product attention is sufficient but not necessary. Without softmax, the approximation capacity of a linearized Transform"},"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":"2105.14995","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-05-31T14:30:53Z","cross_cats_sorted":["cs.NA","math.NA"],"title_canon_sha256":"272b51a9f6d04294b66fa6d9f49509853f4cc7144890aa5a6015170a1e90728a","abstract_canon_sha256":"a44be3bbc815c2a7bbae82096b6073d5f89afbf48a7c0cbfcf2afbe611637a2f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:27:31.807587Z","signature_b64":"ay49zu5aqdo2V9YQ5NyYjNGZkabjfeIt0kpB1baeIB54d6qC4LwKWwEyUJQHilA4PfCWtB5DyL2BVpQ9/LKLBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"409ec1954a06f9df6d773773e349c9756994a87409637f4eedf7c2e290eb1a40","last_reissued_at":"2026-07-05T03:27:31.807136Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:27:31.807136Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Choose a Transformer: Fourier or Galerkin","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NA","math.NA"],"primary_cat":"cs.LG","authors_text":"Shuhao Cao","submitted_at":"2021-05-31T14:30:53Z","abstract_excerpt":"In this paper, we apply the self-attention from the state-of-the-art Transformer in Attention Is All You Need for the first time to a data-driven operator learning problem related to partial differential equations. An effort is put together to explain the heuristics of, and to improve the efficacy of the attention mechanism. By employing the operator approximation theory in Hilbert spaces, it is demonstrated for the first time that the softmax normalization in the scaled dot-product attention is sufficient but not necessary. Without softmax, the approximation capacity of a linearized Transform"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.14995","kind":"arxiv","version":4},"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/2105.14995/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":"2105.14995","created_at":"2026-07-05T03:27:31.807194+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.14995v4","created_at":"2026-07-05T03:27:31.807194+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.14995","created_at":"2026-07-05T03:27:31.807194+00:00"},{"alias_kind":"pith_short_12","alias_value":"ICPMDFKKA345","created_at":"2026-07-05T03:27:31.807194+00:00"},{"alias_kind":"pith_short_16","alias_value":"ICPMDFKKA34563LX","created_at":"2026-07-05T03:27:31.807194+00:00"},{"alias_kind":"pith_short_8","alias_value":"ICPMDFKK","created_at":"2026-07-05T03:27:31.807194+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25569","citing_title":"VesNet: Neural network accelerated solver for simulating Stokesian vesicle suspensions","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18525","citing_title":"Overlapping Schwarz Attention: Hierarchical Attention via Domain Decomposition","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11963","citing_title":"HAMNO: A Hierarchical Adaptive Multi-scale Neural Operator with Physics-Informed Learning for Dynamical Systems","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24876","citing_title":"IV-Net: A neural network for elliptic PDEs with random and highly varying coefficients","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09523","citing_title":"HS-FNO: History-Space Fourier Neural Operator for Non-Markovian Partial Differential Equations","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09523","citing_title":"HS-FNO: History-Space Fourier Neural Operator for Non-Markovian Partial Differential Equations","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06475","citing_title":"AE-ViT: Stable Long-Horizon Parametric Partial Differential Equations Modeling","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV","json":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV.json","graph_json":"https://pith.science/api/pith-number/ICPMDFKKA34563LXG5Z6GSOJOV/graph.json","events_json":"https://pith.science/api/pith-number/ICPMDFKKA34563LXG5Z6GSOJOV/events.json","paper":"https://pith.science/paper/ICPMDFKK"},"agent_actions":{"view_html":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV","download_json":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV.json","view_paper":"https://pith.science/paper/ICPMDFKK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.14995&json=true","fetch_graph":"https://pith.science/api/pith-number/ICPMDFKKA34563LXG5Z6GSOJOV/graph.json","fetch_events":"https://pith.science/api/pith-number/ICPMDFKKA34563LXG5Z6GSOJOV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV/action/storage_attestation","attest_author":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV/action/author_attestation","sign_citation":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV/action/citation_signature","submit_replication":"https://pith.science/pith/ICPMDFKKA34563LXG5Z6GSOJOV/action/replication_record"}},"created_at":"2026-07-05T03:27:31.807194+00:00","updated_at":"2026-07-05T03:27:31.807194+00:00"}