{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:XUQ6RRMXL4AVEDWEWBJH4KNZME","short_pith_number":"pith:XUQ6RRMX","schema_version":"1.0","canonical_sha256":"bd21e8c5975f01520ec4b0527e29b9612589002044dec43df10dd724bdc4663a","source":{"kind":"arxiv","id":"2202.01575","version":3},"attestation_state":"computed","paper":{"title":"CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Akshat Kumar, Chenghao Liu, Doyen Sahoo, Gerald Woo, Steven Hoi","submitted_at":"2022-02-03T13:17:38Z","abstract_excerpt":"Deep learning has been actively studied for time series forecasting, and the mainstream paradigm is based on the end-to-end training of neural network architectures, ranging from classical LSTM/RNNs to more recent TCNs and Transformers. Motivated by the recent success of representation learning in computer vision and natural language processing, we argue that a more promising paradigm for time series forecasting, is to first learn disentangled feature representations, followed by a simple regression fine-tuning step -- we justify such a paradigm from a causal perspective. Following this princi"},"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":"2202.01575","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-02-03T13:17:38Z","cross_cats_sorted":[],"title_canon_sha256":"94454701e8bafd52f2b492fb7f4f30b18dd250b97e790677d6e3ca5e53481eee","abstract_canon_sha256":"4dc9409a7ee40a09482c3141ee34e97b45b4848f87e648915a98e9dafbe6cc91"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:20:32.972779Z","signature_b64":"lmJ2Xp0iiQ4GXv1BOaNr87Wfm5J47mM3kdhhM5p/b8AWENamM/nrd7/EdK/7wgWHIE403oSWZR1NtBqeiQMUBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd21e8c5975f01520ec4b0527e29b9612589002044dec43df10dd724bdc4663a","last_reissued_at":"2026-07-05T04:20:32.972267Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:20:32.972267Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Akshat Kumar, Chenghao Liu, Doyen Sahoo, Gerald Woo, Steven Hoi","submitted_at":"2022-02-03T13:17:38Z","abstract_excerpt":"Deep learning has been actively studied for time series forecasting, and the mainstream paradigm is based on the end-to-end training of neural network architectures, ranging from classical LSTM/RNNs to more recent TCNs and Transformers. Motivated by the recent success of representation learning in computer vision and natural language processing, we argue that a more promising paradigm for time series forecasting, is to first learn disentangled feature representations, followed by a simple regression fine-tuning step -- we justify such a paradigm from a causal perspective. Following this princi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2202.01575","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/2202.01575/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":"2202.01575","created_at":"2026-07-05T04:20:32.972338+00:00"},{"alias_kind":"arxiv_version","alias_value":"2202.01575v3","created_at":"2026-07-05T04:20:32.972338+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2202.01575","created_at":"2026-07-05T04:20:32.972338+00:00"},{"alias_kind":"pith_short_12","alias_value":"XUQ6RRMXL4AV","created_at":"2026-07-05T04:20:32.972338+00:00"},{"alias_kind":"pith_short_16","alias_value":"XUQ6RRMXL4AVEDWE","created_at":"2026-07-05T04:20:32.972338+00:00"},{"alias_kind":"pith_short_8","alias_value":"XUQ6RRMX","created_at":"2026-07-05T04:20:32.972338+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01918","citing_title":"Zeus: Towards Tuning-Free Foundation Model for Time Series Analysis","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10592","citing_title":"Dirichlet-Guided Group Forecasting for Alleviating Over-smoothing in Time Series Forecasting","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00958","citing_title":"LeNEPA: No-Augmentation Next-Latent Prediction for Time-Series Representation Learning","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00956","citing_title":"Aionoscope: Debugging Latent-State Accessibility in Time-Series Representations","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22055","citing_title":"Prototype-Guided Classification Sub-Task Decoupling Framework: Enhancing Generalization and Interpretability for Multivariate Time Series","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2512.16001","citing_title":"Concurrence: A dependence criterion for time series, applied to biological data","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10291","citing_title":"TimeSeriesExamAgent: Creating Time Series Reasoning Benchmarks at Scale","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME","json":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME.json","graph_json":"https://pith.science/api/pith-number/XUQ6RRMXL4AVEDWEWBJH4KNZME/graph.json","events_json":"https://pith.science/api/pith-number/XUQ6RRMXL4AVEDWEWBJH4KNZME/events.json","paper":"https://pith.science/paper/XUQ6RRMX"},"agent_actions":{"view_html":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME","download_json":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME.json","view_paper":"https://pith.science/paper/XUQ6RRMX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2202.01575&json=true","fetch_graph":"https://pith.science/api/pith-number/XUQ6RRMXL4AVEDWEWBJH4KNZME/graph.json","fetch_events":"https://pith.science/api/pith-number/XUQ6RRMXL4AVEDWEWBJH4KNZME/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME/action/storage_attestation","attest_author":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME/action/author_attestation","sign_citation":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME/action/citation_signature","submit_replication":"https://pith.science/pith/XUQ6RRMXL4AVEDWEWBJH4KNZME/action/replication_record"}},"created_at":"2026-07-05T04:20:32.972338+00:00","updated_at":"2026-07-05T04:20:32.972338+00:00"}