{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:K4ON45AVXF4FXQACX5DKJVL7XI","short_pith_number":"pith:K4ON45AV","schema_version":"1.0","canonical_sha256":"571cde7415b9785bc002bf46a4d57fba1b41d2daf38cfe74d1c462f4f8b36669","source":{"kind":"arxiv","id":"2110.02642","version":5},"attestation_state":"computed","paper":{"title":"Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haixu Wu, Jianmin Wang, Jiehui Xu, Mingsheng Long","submitted_at":"2021-10-06T10:33:55Z","abstract_excerpt":"Unsupervised detection of anomaly points in time series is a challenging problem, which requires the model to derive a distinguishable criterion. Previous methods tackle the problem mainly through learning pointwise representation or pairwise association, however, neither is sufficient to reason about the intricate dynamics. Recently, Transformers have shown great power in unified modeling of pointwise representation and pairwise association, and we find that the self-attention weight distribution of each time point can embody rich association with the whole series. Our key observation is that"},"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":"2110.02642","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2021-10-06T10:33:55Z","cross_cats_sorted":[],"title_canon_sha256":"f721027a833aee3bbe2837323f87c68fa0b0e6e85c392b1e9aa0dec62eaa7db0","abstract_canon_sha256":"8950823f41a2e56f86f79ef4ea95fe88e51fa2d8af1d6236ddc7eed5322e2e4e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:35:57.066711Z","signature_b64":"4flrMrdbnj/FXNnJeBre/muCXkVPZwGYYM81Nmf0kyXQzI5nGhV9ZPAMsEFQ3N7zl8yomtyFdkKlJzLR9XItBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"571cde7415b9785bc002bf46a4d57fba1b41d2daf38cfe74d1c462f4f8b36669","last_reissued_at":"2026-07-05T04:35:57.066160Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:35:57.066160Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Haixu Wu, Jianmin Wang, Jiehui Xu, Mingsheng Long","submitted_at":"2021-10-06T10:33:55Z","abstract_excerpt":"Unsupervised detection of anomaly points in time series is a challenging problem, which requires the model to derive a distinguishable criterion. Previous methods tackle the problem mainly through learning pointwise representation or pairwise association, however, neither is sufficient to reason about the intricate dynamics. Recently, Transformers have shown great power in unified modeling of pointwise representation and pairwise association, and we find that the self-attention weight distribution of each time point can embody rich association with the whole series. Our key observation is that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.02642","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/2110.02642/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":"2110.02642","created_at":"2026-07-05T04:35:57.066223+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.02642v5","created_at":"2026-07-05T04:35:57.066223+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.02642","created_at":"2026-07-05T04:35:57.066223+00:00"},{"alias_kind":"pith_short_12","alias_value":"K4ON45AVXF4F","created_at":"2026-07-05T04:35:57.066223+00:00"},{"alias_kind":"pith_short_16","alias_value":"K4ON45AVXF4FXQAC","created_at":"2026-07-05T04:35:57.066223+00:00"},{"alias_kind":"pith_short_8","alias_value":"K4ON45AV","created_at":"2026-07-05T04:35:57.066223+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05805","citing_title":"Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2607.01918","citing_title":"Zeus: Towards Tuning-Free Foundation Model for Time Series Analysis","ref_index":95,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02046","citing_title":"Fast and Accurate Anomaly Detection in Time Series","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00720","citing_title":"Detecting the Undetectable: Enhancing Unsupervised time series Anomaly Detection via Active Learning","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09874","citing_title":"Disjoint or Overlapping? Inference Windowing for Reconstruction-Based Time Series Anomaly Detection","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27486","citing_title":"Federated Learning for Multivariate Time Series Anomaly Detection in Industrial Automation","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00304","citing_title":"Modeling Spectral Energy Shifts in Spatio-Temporal Graph Anomaly Detection","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2410.04047","citing_title":"TS-Reasoner: Domain-Oriented Time Series Inference Agents for Reasoning and Automated Analysis","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2507.15066","citing_title":"Time-RA: Towards Time Series Reasoning for Anomaly Diagnosis with LLM Feedback","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2511.15083","citing_title":"Fourier-KAN-Mamba: A Novel State-Space Equation Approach for Time-Series Anomaly Detection","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2512.12116","citing_title":"Neural CDEs as Correctors for Learned Time Series Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21293","citing_title":"Physics-Guided Tiny-Mamba Transformer for Reliability-Aware Early Fault Warning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08582","citing_title":"Multivariate Time Series Anomaly Detection via Dual-Branch Reconstruction and Autoregressive Flow-based Residual Density Estimation","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09685","citing_title":"Learning Unified Representations of Normalcy for Time Series Anomaly Detection","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21529","citing_title":"Architectures for Robust Self-Organizing Energy Systems under Information and Control Constraints","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17998","citing_title":"Causally-Constrained Probabilistic Forecasting for Time-Series Anomaly Detection","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17616","citing_title":"Conditional Attribution for Root Cause Analysis in Time-Series Anomaly Detection","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI","json":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI.json","graph_json":"https://pith.science/api/pith-number/K4ON45AVXF4FXQACX5DKJVL7XI/graph.json","events_json":"https://pith.science/api/pith-number/K4ON45AVXF4FXQACX5DKJVL7XI/events.json","paper":"https://pith.science/paper/K4ON45AV"},"agent_actions":{"view_html":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI","download_json":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI.json","view_paper":"https://pith.science/paper/K4ON45AV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.02642&json=true","fetch_graph":"https://pith.science/api/pith-number/K4ON45AVXF4FXQACX5DKJVL7XI/graph.json","fetch_events":"https://pith.science/api/pith-number/K4ON45AVXF4FXQACX5DKJVL7XI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI/action/storage_attestation","attest_author":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI/action/author_attestation","sign_citation":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI/action/citation_signature","submit_replication":"https://pith.science/pith/K4ON45AVXF4FXQACX5DKJVL7XI/action/replication_record"}},"created_at":"2026-07-05T04:35:57.066223+00:00","updated_at":"2026-07-05T04:35:57.066223+00:00"}