{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BIN2ZJL2JUG2PAKDCXWYE5LJ3E","short_pith_number":"pith:BIN2ZJL2","schema_version":"1.0","canonical_sha256":"0a1baca57a4d0da7814315ed827569d9080db00689188a8bf5b10614aed96b69","source":{"kind":"arxiv","id":"2604.16429","version":3},"attestation_state":"computed","paper":{"title":"(Sparse) Attention to the Details: Preserving Spectral Fidelity in ML-based Weather Forecasting Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Mosaic achieves near-perfect spectral alignment in 1.5° weather forecasts by using block-sparse attention and learned ensemble perturbations, matching finer-resolution models.","cross_cats":["cs.AI","cs.CV","physics.ao-ph"],"primary_cat":"cs.LG","authors_text":"Ana Lucic, Jan-Willem van de Meent, Maksim Zhdanov, Max Welling","submitted_at":"2026-04-06T08:50:42Z","abstract_excerpt":"We introduce Mosaic, a probabilistic weather forecasting model that addresses three failure modes of spectral degradation in ML-based weather prediction: spectral damping (statistical), high-frequency aliasing (architectural), and residual high-frequency leakage (parametric). Mosaic generates ensemble members through learned functional perturbations and operates on native-resolution grids via mesh-aligned block-sparse attention, a hardware-aligned mechanism that captures long-range dependencies at linear cost by sharing keys and values across spatially adjacent queries. At 1.5{\\deg} resolution"},"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":true},"canonical_record":{"source":{"id":"2604.16429","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-06T08:50:42Z","cross_cats_sorted":["cs.AI","cs.CV","physics.ao-ph"],"title_canon_sha256":"c8538bfff0f2835dfb25b2f4bbf501ea5ac8b38070aa6e7e41a6768c4998c663","abstract_canon_sha256":"59a59c77216e980dc0f0406b6986c4e202c6a6f9fccd0ea129574f1a308c15c3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:31.916695Z","signature_b64":"8AHDb2xjUDcm289WEwBurpU6FIxVXhOeG9frp9wKY+CNj8VgU784tt0uOxuxYwBW5GIKfHL2ovtjLWKyqjQqCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a1baca57a4d0da7814315ed827569d9080db00689188a8bf5b10614aed96b69","last_reissued_at":"2026-05-20T00:04:31.915899Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:31.915899Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"(Sparse) Attention to the Details: Preserving Spectral Fidelity in ML-based Weather Forecasting Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Mosaic achieves near-perfect spectral alignment in 1.5° weather forecasts by using block-sparse attention and learned ensemble perturbations, matching finer-resolution models.","cross_cats":["cs.AI","cs.CV","physics.ao-ph"],"primary_cat":"cs.LG","authors_text":"Ana Lucic, Jan-Willem van de Meent, Maksim Zhdanov, Max Welling","submitted_at":"2026-04-06T08:50:42Z","abstract_excerpt":"We introduce Mosaic, a probabilistic weather forecasting model that addresses three failure modes of spectral degradation in ML-based weather prediction: spectral damping (statistical), high-frequency aliasing (architectural), and residual high-frequency leakage (parametric). Mosaic generates ensemble members through learned functional perturbations and operates on native-resolution grids via mesh-aligned block-sparse attention, a hardware-aligned mechanism that captures long-range dependencies at linear cost by sharing keys and values across spatially adjacent queries. At 1.5{\\deg} resolution"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Mosaic produces well-calibrated ensembles whose individual members exhibit near-perfect spectral alignment across all resolved frequencies at 1.5° resolution while matching or outperforming models trained on 6× finer grids.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the mesh-aligned block-sparse attention fully captures necessary long-range dependencies without introducing new artifacts or losing critical interactions that standard attention would preserve.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Mosaic achieves state-of-the-art spectral alignment in 1.5° weather forecasts via learned functional perturbations and hardware-aligned sparse attention, matching finer-resolution models with fast inference.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Mosaic achieves near-perfect spectral alignment in 1.5° weather forecasts by using block-sparse attention and learned ensemble perturbations, matching finer-resolution models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d1ea6d4dafd37b86feaba14960ed678d8150e8bed21341459835f0c385df26a5"},"source":{"id":"2604.16429","kind":"arxiv","version":3},"verdict":{"id":"cd44595f-8101-42c0-980a-4cbec0767b67","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:13:35.022788Z","strongest_claim":"Mosaic produces well-calibrated ensembles whose individual members exhibit near-perfect spectral alignment across all resolved frequencies at 1.5° resolution while matching or outperforming models trained on 6× finer grids.","one_line_summary":"Mosaic achieves state-of-the-art spectral alignment in 1.5° weather forecasts via learned functional perturbations and hardware-aligned sparse attention, matching finer-resolution models with fast inference.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the mesh-aligned block-sparse attention fully captures necessary long-range dependencies without introducing new artifacts or losing critical interactions that standard attention would preserve.","pith_extraction_headline":"Mosaic achieves near-perfect spectral alignment in 1.5° weather forecasts by using block-sparse attention and learned ensemble perturbations, matching finer-resolution models."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.16429/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":2,"snapshot_sha256":"0d4873aeb516f38615bdd0522eebdc9cd70bfd65afef0f54aaf12ac19664c4d5"},"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":"2604.16429","created_at":"2026-05-20T00:04:31.916026+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.16429v3","created_at":"2026-05-20T00:04:31.916026+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.16429","created_at":"2026-05-20T00:04:31.916026+00:00"},{"alias_kind":"pith_short_12","alias_value":"BIN2ZJL2JUG2","created_at":"2026-05-20T00:04:31.916026+00:00"},{"alias_kind":"pith_short_16","alias_value":"BIN2ZJL2JUG2PAKD","created_at":"2026-05-20T00:04:31.916026+00:00"},{"alias_kind":"pith_short_8","alias_value":"BIN2ZJL2","created_at":"2026-05-20T00:04:31.916026+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.19939","citing_title":"Uncertainty-aware Machine Learning Interatomic Potentials via Learned Functional Perturbations","ref_index":2,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E","json":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E.json","graph_json":"https://pith.science/api/pith-number/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/graph.json","events_json":"https://pith.science/api/pith-number/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/events.json","paper":"https://pith.science/paper/BIN2ZJL2"},"agent_actions":{"view_html":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E","download_json":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E.json","view_paper":"https://pith.science/paper/BIN2ZJL2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.16429&json=true","fetch_graph":"https://pith.science/api/pith-number/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/graph.json","fetch_events":"https://pith.science/api/pith-number/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/action/storage_attestation","attest_author":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/action/author_attestation","sign_citation":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/action/citation_signature","submit_replication":"https://pith.science/pith/BIN2ZJL2JUG2PAKDCXWYE5LJ3E/action/replication_record"}},"created_at":"2026-05-20T00:04:31.916026+00:00","updated_at":"2026-05-20T00:04:31.916026+00:00"}