{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WWMISSWEBKG566BHE7CZCSLWDD","short_pith_number":"pith:WWMISSWE","schema_version":"1.0","canonical_sha256":"b598894ac40a8ddf782727c591497618d7caf7ef771281a78658d116d2623c56","source":{"kind":"arxiv","id":"2402.02332","version":3},"attestation_state":"computed","paper":{"title":"Minusformer: Improving Time Series Forecasting by Progressively Learning Residuals","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bingzheng Zhang, Daojun Liang, Dongfeng Yuan, Haixia Zhang, Minggao Zhang","submitted_at":"2024-02-04T03:54:31Z","abstract_excerpt":"In this paper, we find that ubiquitous time series (TS) forecasting models are prone to severe overfitting. To cope with this problem, we embrace a de-redundancy approach to progressively reinstate the intrinsic values of TS for future intervals. Specifically, we introduce a dual-stream and subtraction mechanism, which is a deep Boosting ensemble learning method. And the vanilla Transformer is renovated by reorienting the information aggregation mechanism from addition to subtraction. Then, we incorporate an auxiliary output branch into each block of the original model to construct a highway l"},"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":"2402.02332","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-04T03:54:31Z","cross_cats_sorted":[],"title_canon_sha256":"0e330f4ba36a0d8388d53628e3e186de7836efbed8f261b1861d73008b8071f2","abstract_canon_sha256":"b3342c31b951b71f3cff6435074fade7e2e599ba87e3370159e3b527ebad7561"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:32:35.800102Z","signature_b64":"DdCCOIvBoxwF4mAP1x9PU4nUBTzs/w/fcPx/I4TFRYTsd/k3ZdIKasz1ifTGC1tZjYcelEe3b3Ijd8ZcpGfzAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b598894ac40a8ddf782727c591497618d7caf7ef771281a78658d116d2623c56","last_reissued_at":"2026-07-05T08:32:35.799524Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:32:35.799524Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Minusformer: Improving Time Series Forecasting by Progressively Learning Residuals","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bingzheng Zhang, Daojun Liang, Dongfeng Yuan, Haixia Zhang, Minggao Zhang","submitted_at":"2024-02-04T03:54:31Z","abstract_excerpt":"In this paper, we find that ubiquitous time series (TS) forecasting models are prone to severe overfitting. To cope with this problem, we embrace a de-redundancy approach to progressively reinstate the intrinsic values of TS for future intervals. Specifically, we introduce a dual-stream and subtraction mechanism, which is a deep Boosting ensemble learning method. And the vanilla Transformer is renovated by reorienting the information aggregation mechanism from addition to subtraction. Then, we incorporate an auxiliary output branch into each block of the original model to construct a highway l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.02332","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/2402.02332/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":"2402.02332","created_at":"2026-07-05T08:32:35.799594+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.02332v3","created_at":"2026-07-05T08:32:35.799594+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.02332","created_at":"2026-07-05T08:32:35.799594+00:00"},{"alias_kind":"pith_short_12","alias_value":"WWMISSWEBKG5","created_at":"2026-07-05T08:32:35.799594+00:00"},{"alias_kind":"pith_short_16","alias_value":"WWMISSWEBKG566BH","created_at":"2026-07-05T08:32:35.799594+00:00"},{"alias_kind":"pith_short_8","alias_value":"WWMISSWE","created_at":"2026-07-05T08:32:35.799594+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.23116","citing_title":"CrossLinear: Plug-and-Play Cross-Correlation Embedding for Time Series Forecasting with Exogenous Variables","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD","json":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD.json","graph_json":"https://pith.science/api/pith-number/WWMISSWEBKG566BHE7CZCSLWDD/graph.json","events_json":"https://pith.science/api/pith-number/WWMISSWEBKG566BHE7CZCSLWDD/events.json","paper":"https://pith.science/paper/WWMISSWE"},"agent_actions":{"view_html":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD","download_json":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD.json","view_paper":"https://pith.science/paper/WWMISSWE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.02332&json=true","fetch_graph":"https://pith.science/api/pith-number/WWMISSWEBKG566BHE7CZCSLWDD/graph.json","fetch_events":"https://pith.science/api/pith-number/WWMISSWEBKG566BHE7CZCSLWDD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD/action/storage_attestation","attest_author":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD/action/author_attestation","sign_citation":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD/action/citation_signature","submit_replication":"https://pith.science/pith/WWMISSWEBKG566BHE7CZCSLWDD/action/replication_record"}},"created_at":"2026-07-05T08:32:35.799594+00:00","updated_at":"2026-07-05T08:32:35.799594+00:00"}