{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:W7ZBT67ZQS5EP2LRO7F4RPIPIR","short_pith_number":"pith:W7ZBT67Z","schema_version":"1.0","canonical_sha256":"b7f219fbf984ba47e97177cbc8bd0f4473829239a86c310d16fef9ad3f020ecf","source":{"kind":"arxiv","id":"2607.21080","version":1},"attestation_state":"computed","paper":{"title":"Nipping the Butterfly Effect in the Bud: Self-Output Fine-Tuning for Autoregressive Weather Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hsuan-Tien Lin, Yun-Ye Cai","submitted_at":"2026-07-23T09:10:11Z","abstract_excerpt":"Long-horizon weather forecasting is a fundamental challenge in atmospheric science, for which autoregressive Deep Learning Weather Prediction (DLWP) has emerged as the primary paradigm. Although the autoregressive pipeline is highly scalable and flexible, its prediction errors grow rapidly over long forecasting horizons. In this work, we study this error growth phenomenon from both theoretical and empirical perspectives. Our analysis reveals that the growth is driven by a feedback loop between output errors and input distribution shifts. Specifically, the autoregressive process amplifies small"},"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":"2607.21080","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-23T09:10:11Z","cross_cats_sorted":[],"title_canon_sha256":"11ef404120fa55b0ab78cdab94af007ae946a00d3373548ffd060d1204627489","abstract_canon_sha256":"c31734ae0e8e6727d5e4c5fb8c94282eef59732a98f20b89b96d14ee6dc40c3a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T01:23:50.864271Z","signature_b64":"GduKufIDEg0q6OKEB1ZZH6WhIZJBaP8VHbbiYNbiold0WwcwRzWhuaEV3ayKJcwcgFXGrykoMdMeqbN8p0gvAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b7f219fbf984ba47e97177cbc8bd0f4473829239a86c310d16fef9ad3f020ecf","last_reissued_at":"2026-07-24T01:23:50.863434Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T01:23:50.863434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Nipping the Butterfly Effect in the Bud: Self-Output Fine-Tuning for Autoregressive Weather Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Hsuan-Tien Lin, Yun-Ye Cai","submitted_at":"2026-07-23T09:10:11Z","abstract_excerpt":"Long-horizon weather forecasting is a fundamental challenge in atmospheric science, for which autoregressive Deep Learning Weather Prediction (DLWP) has emerged as the primary paradigm. Although the autoregressive pipeline is highly scalable and flexible, its prediction errors grow rapidly over long forecasting horizons. In this work, we study this error growth phenomenon from both theoretical and empirical perspectives. Our analysis reveals that the growth is driven by a feedback loop between output errors and input distribution shifts. Specifically, the autoregressive process amplifies small"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21080","kind":"arxiv","version":1},"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/2607.21080/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":"2607.21080","created_at":"2026-07-24T01:23:50.863866+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.21080v1","created_at":"2026-07-24T01:23:50.863866+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21080","created_at":"2026-07-24T01:23:50.863866+00:00"},{"alias_kind":"pith_short_12","alias_value":"W7ZBT67ZQS5E","created_at":"2026-07-24T01:23:50.863866+00:00"},{"alias_kind":"pith_short_16","alias_value":"W7ZBT67ZQS5EP2LR","created_at":"2026-07-24T01:23:50.863866+00:00"},{"alias_kind":"pith_short_8","alias_value":"W7ZBT67Z","created_at":"2026-07-24T01:23:50.863866+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR","json":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR.json","graph_json":"https://pith.science/api/pith-number/W7ZBT67ZQS5EP2LRO7F4RPIPIR/graph.json","events_json":"https://pith.science/api/pith-number/W7ZBT67ZQS5EP2LRO7F4RPIPIR/events.json","paper":"https://pith.science/paper/W7ZBT67Z"},"agent_actions":{"view_html":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR","download_json":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR.json","view_paper":"https://pith.science/paper/W7ZBT67Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.21080&json=true","fetch_graph":"https://pith.science/api/pith-number/W7ZBT67ZQS5EP2LRO7F4RPIPIR/graph.json","fetch_events":"https://pith.science/api/pith-number/W7ZBT67ZQS5EP2LRO7F4RPIPIR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR/action/storage_attestation","attest_author":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR/action/author_attestation","sign_citation":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR/action/citation_signature","submit_replication":"https://pith.science/pith/W7ZBT67ZQS5EP2LRO7F4RPIPIR/action/replication_record"}},"created_at":"2026-07-24T01:23:50.863866+00:00","updated_at":"2026-07-24T01:23:50.863866+00:00"}