{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:GAOJJYO2IENAP26C3L6B6LHNVT","short_pith_number":"pith:GAOJJYO2","schema_version":"1.0","canonical_sha256":"301c94e1da411a07ebc2dafc1f2cedacda53ddef0d57c77ae78b705f2d8c7cd5","source":{"kind":"arxiv","id":"2410.05711","version":5},"attestation_state":"computed","paper":{"title":"TimeDART: A Diffusion Autoregressive Transformer for Self-Supervised Time Series Representation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daoyu Wang, Mingyue Cheng, Qi Liu, Zhiding Liu","submitted_at":"2024-10-08T06:08:33Z","abstract_excerpt":"Self-supervised learning has garnered increasing attention in time series analysis for benefiting various downstream tasks and reducing reliance on labeled data. Despite its effectiveness, existing methods often struggle to comprehensively capture both long-term dynamic evolution and subtle local patterns in a unified manner. In this work, we propose \\textbf{TimeDART}, a novel self-supervised time series pre-training framework that unifies two powerful generative paradigms to learn more transferable representations. Specifically, we first employ a causal Transformer encoder, accompanied by a p"},"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":"2410.05711","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-08T06:08:33Z","cross_cats_sorted":[],"title_canon_sha256":"1822c053e3a0111b2531be8d951b4d93db005af45e825e12d0d39848cc83ff08","abstract_canon_sha256":"87b7afcb477ddcd174e1a5b84f7a8b3bb09ea25e00983bc4b589bfe2b4545290"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:21.560542Z","signature_b64":"AHYQjwFvEW5RXeZBkisOf/XymKjTkRWmjOa6UFicvNw+ctF0/GI4SS8d047uhHgHmga15j9Ao5Z4pfRLAE8gCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"301c94e1da411a07ebc2dafc1f2cedacda53ddef0d57c77ae78b705f2d8c7cd5","last_reissued_at":"2026-07-05T11:19:21.560022Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:21.560022Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TimeDART: A Diffusion Autoregressive Transformer for Self-Supervised Time Series Representation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Daoyu Wang, Mingyue Cheng, Qi Liu, Zhiding Liu","submitted_at":"2024-10-08T06:08:33Z","abstract_excerpt":"Self-supervised learning has garnered increasing attention in time series analysis for benefiting various downstream tasks and reducing reliance on labeled data. Despite its effectiveness, existing methods often struggle to comprehensively capture both long-term dynamic evolution and subtle local patterns in a unified manner. In this work, we propose \\textbf{TimeDART}, a novel self-supervised time series pre-training framework that unifies two powerful generative paradigms to learn more transferable representations. Specifically, we first employ a causal Transformer encoder, accompanied by a p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05711","kind":"arxiv","version":5},"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/2410.05711/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":"2410.05711","created_at":"2026-07-05T11:19:21.560090+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.05711v5","created_at":"2026-07-05T11:19:21.560090+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05711","created_at":"2026-07-05T11:19:21.560090+00:00"},{"alias_kind":"pith_short_12","alias_value":"GAOJJYO2IENA","created_at":"2026-07-05T11:19:21.560090+00:00"},{"alias_kind":"pith_short_16","alias_value":"GAOJJYO2IENAP26C","created_at":"2026-07-05T11:19:21.560090+00:00"},{"alias_kind":"pith_short_8","alias_value":"GAOJJYO2","created_at":"2026-07-05T11:19:21.560090+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02138","citing_title":"VLBM: Variational Latent Basis Modeling for OOD Robust Multivariate Time Series Forecasting","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28446","citing_title":"Domain-Informed Multi-View Self-Distillation for Astronomical Light-Curve Representation Learning with JEPA","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19462","citing_title":"Quantifying the Pre-training Dividend: Generative versus Latent Self-Supervised Learning for Time Series Foundation Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06727","citing_title":"Bi-level Heterogeneous Learning for Time Series Foundation Models: A Federated Learning Approach","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT","json":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT.json","graph_json":"https://pith.science/api/pith-number/GAOJJYO2IENAP26C3L6B6LHNVT/graph.json","events_json":"https://pith.science/api/pith-number/GAOJJYO2IENAP26C3L6B6LHNVT/events.json","paper":"https://pith.science/paper/GAOJJYO2"},"agent_actions":{"view_html":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT","download_json":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT.json","view_paper":"https://pith.science/paper/GAOJJYO2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.05711&json=true","fetch_graph":"https://pith.science/api/pith-number/GAOJJYO2IENAP26C3L6B6LHNVT/graph.json","fetch_events":"https://pith.science/api/pith-number/GAOJJYO2IENAP26C3L6B6LHNVT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT/action/storage_attestation","attest_author":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT/action/author_attestation","sign_citation":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT/action/citation_signature","submit_replication":"https://pith.science/pith/GAOJJYO2IENAP26C3L6B6LHNVT/action/replication_record"}},"created_at":"2026-07-05T11:19:21.560090+00:00","updated_at":"2026-07-05T11:19:21.560090+00:00"}