{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HZR37Z7UN2XYNDRNFHG3RSCEIO","short_pith_number":"pith:HZR37Z7U","schema_version":"1.0","canonical_sha256":"3e63bfe7f46eaf868e2d29cdb8c84443897ddd5200ee9971b3b5fc0980e34fee","source":{"kind":"arxiv","id":"2506.23424","version":1},"attestation_state":"computed","paper":{"title":"Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecasting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Gabriel L. Oliveira, Heitor R. Medeiros, Hossein Sharifi-Noghabi, Saghar Irandoust","submitted_at":"2025-06-29T23:09:35Z","abstract_excerpt":"Real-world time series often exhibit a non-stationary nature, degrading the performance of pre-trained forecasting models. Test-Time Adaptation (TTA) addresses this by adjusting models during inference, but existing methods typically update the full model, increasing memory and compute costs. We propose PETSA, a parameter-efficient method that adapts forecasters at test time by only updating small calibration modules on the input and output. PETSA uses low-rank adapters and dynamic gating to adjust representations without retraining. To maintain accuracy despite limited adaptation capacity, we"},"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":"2506.23424","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-29T23:09:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"357d40eecf8e23c12a2c899d897979b7e5b5a5ea914b220be0a3c5320d375370","abstract_canon_sha256":"0c07b687188fec48316409f62cf21db119fd1346119cd93d0fa92622a99da14c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:29:21.430342Z","signature_b64":"gxSkJBXc0XeJ2Vsn5z/DTGOnLlUWGY7vc72u6R8RhoJVmBRdEKd/KZ7dndPT2yZ5kox6bZleEuO2MYq42ly8AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e63bfe7f46eaf868e2d29cdb8c84443897ddd5200ee9971b3b5fc0980e34fee","last_reissued_at":"2026-07-05T11:29:21.429857Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:29:21.429857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecasting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Gabriel L. Oliveira, Heitor R. Medeiros, Hossein Sharifi-Noghabi, Saghar Irandoust","submitted_at":"2025-06-29T23:09:35Z","abstract_excerpt":"Real-world time series often exhibit a non-stationary nature, degrading the performance of pre-trained forecasting models. Test-Time Adaptation (TTA) addresses this by adjusting models during inference, but existing methods typically update the full model, increasing memory and compute costs. We propose PETSA, a parameter-efficient method that adapts forecasters at test time by only updating small calibration modules on the input and output. PETSA uses low-rank adapters and dynamic gating to adjust representations without retraining. To maintain accuracy despite limited adaptation capacity, we"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23424","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/2506.23424/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":"2506.23424","created_at":"2026-07-05T11:29:21.429913+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.23424v1","created_at":"2026-07-05T11:29:21.429913+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23424","created_at":"2026-07-05T11:29:21.429913+00:00"},{"alias_kind":"pith_short_12","alias_value":"HZR37Z7UN2XY","created_at":"2026-07-05T11:29:21.429913+00:00"},{"alias_kind":"pith_short_16","alias_value":"HZR37Z7UN2XYNDRN","created_at":"2026-07-05T11:29:21.429913+00:00"},{"alias_kind":"pith_short_8","alias_value":"HZR37Z7U","created_at":"2026-07-05T11:29:21.429913+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25068","citing_title":"Adapt Only When It Pays: Budgeted Decision-Loss Priority for Delayed Online Time-Series Adaptation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08005","citing_title":"STEPS: A Temporal Smooth Error Propagation Solver on the Manifolds for Test-Time Adaptation in Time Series Forecasting","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO","json":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO.json","graph_json":"https://pith.science/api/pith-number/HZR37Z7UN2XYNDRNFHG3RSCEIO/graph.json","events_json":"https://pith.science/api/pith-number/HZR37Z7UN2XYNDRNFHG3RSCEIO/events.json","paper":"https://pith.science/paper/HZR37Z7U"},"agent_actions":{"view_html":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO","download_json":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO.json","view_paper":"https://pith.science/paper/HZR37Z7U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.23424&json=true","fetch_graph":"https://pith.science/api/pith-number/HZR37Z7UN2XYNDRNFHG3RSCEIO/graph.json","fetch_events":"https://pith.science/api/pith-number/HZR37Z7UN2XYNDRNFHG3RSCEIO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO/action/storage_attestation","attest_author":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO/action/author_attestation","sign_citation":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO/action/citation_signature","submit_replication":"https://pith.science/pith/HZR37Z7UN2XYNDRNFHG3RSCEIO/action/replication_record"}},"created_at":"2026-07-05T11:29:21.429913+00:00","updated_at":"2026-07-05T11:29:21.429913+00:00"}