{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QB7IYXSU47BWPU6MY7RJMSW4HN","short_pith_number":"pith:QB7IYXSU","schema_version":"1.0","canonical_sha256":"807e8c5e54e7c367d3ccc7e2964adc3b6eb4d27b2a93cc3f1c495fc0f0d7edf0","source":{"kind":"arxiv","id":"2106.13924","version":2},"attestation_state":"computed","paper":{"title":"Self-Attentive Ensemble Transformer: Representing Ensemble Interactions in Neural Networks for Earth System Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["physics.ao-ph"],"primary_cat":"cs.LG","authors_text":"Tobias Sebastian Finn","submitted_at":"2021-06-21T18:17:34Z","abstract_excerpt":"Ensemble data from Earth system models has to be calibrated and post-processed. I propose a novel member-by-member post-processing approach with neural networks. I bridge ideas from ensemble data assimilation with self-attention, resulting into the self-attentive ensemble transformer. Here, interactions between ensemble members are represented as additive and dynamic self-attentive part. As proof-of-concept, I regress global ECMWF ensemble forecasts to 2-metre-temperature fields from the ERA5 reanalysis. I demonstrate that the ensemble transformer can calibrate the ensemble spread and extract "},"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":"2106.13924","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-21T18:17:34Z","cross_cats_sorted":["physics.ao-ph"],"title_canon_sha256":"8543242c3762e780ceda0f998262306475fcd0780dd599d8c30dbd2603cd0e3c","abstract_canon_sha256":"3613ce7772e48d7902c20ae04331b7b0dea8fa18cd52a0ea3115aab9a20709d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:56:40.330723Z","signature_b64":"t1soCMN1QIbAgT7TrHX//kv/fijcvouHzKvbqQyYzmFHhRKgWsLB6q0f3iLX1PvlyEKfL5G1NZZ/v5h6I6M1Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"807e8c5e54e7c367d3ccc7e2964adc3b6eb4d27b2a93cc3f1c495fc0f0d7edf0","last_reissued_at":"2026-07-05T02:56:40.330244Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:56:40.330244Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Attentive Ensemble Transformer: Representing Ensemble Interactions in Neural Networks for Earth System Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["physics.ao-ph"],"primary_cat":"cs.LG","authors_text":"Tobias Sebastian Finn","submitted_at":"2021-06-21T18:17:34Z","abstract_excerpt":"Ensemble data from Earth system models has to be calibrated and post-processed. I propose a novel member-by-member post-processing approach with neural networks. I bridge ideas from ensemble data assimilation with self-attention, resulting into the self-attentive ensemble transformer. Here, interactions between ensemble members are represented as additive and dynamic self-attentive part. As proof-of-concept, I regress global ECMWF ensemble forecasts to 2-metre-temperature fields from the ERA5 reanalysis. I demonstrate that the ensemble transformer can calibrate the ensemble spread and extract "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.13924","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/2106.13924/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":"2106.13924","created_at":"2026-07-05T02:56:40.330307+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.13924v2","created_at":"2026-07-05T02:56:40.330307+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.13924","created_at":"2026-07-05T02:56:40.330307+00:00"},{"alias_kind":"pith_short_12","alias_value":"QB7IYXSU47BW","created_at":"2026-07-05T02:56:40.330307+00:00"},{"alias_kind":"pith_short_16","alias_value":"QB7IYXSU47BWPU6M","created_at":"2026-07-05T02:56:40.330307+00:00"},{"alias_kind":"pith_short_8","alias_value":"QB7IYXSU","created_at":"2026-07-05T02:56:40.330307+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.16238","citing_title":"Enhancing AI and Dynamical Subseasonal Forecasts with Probabilistic Bias Correction","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN","json":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN.json","graph_json":"https://pith.science/api/pith-number/QB7IYXSU47BWPU6MY7RJMSW4HN/graph.json","events_json":"https://pith.science/api/pith-number/QB7IYXSU47BWPU6MY7RJMSW4HN/events.json","paper":"https://pith.science/paper/QB7IYXSU"},"agent_actions":{"view_html":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN","download_json":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN.json","view_paper":"https://pith.science/paper/QB7IYXSU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.13924&json=true","fetch_graph":"https://pith.science/api/pith-number/QB7IYXSU47BWPU6MY7RJMSW4HN/graph.json","fetch_events":"https://pith.science/api/pith-number/QB7IYXSU47BWPU6MY7RJMSW4HN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN/action/storage_attestation","attest_author":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN/action/author_attestation","sign_citation":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN/action/citation_signature","submit_replication":"https://pith.science/pith/QB7IYXSU47BWPU6MY7RJMSW4HN/action/replication_record"}},"created_at":"2026-07-05T02:56:40.330307+00:00","updated_at":"2026-07-05T02:56:40.330307+00:00"}