{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.18539", "kind": "paper", "url": "https://arxiv.org/abs/2606.18539"}, {"id": "gh:ray-zyy/ts-fault", "kind": "repo", "url": "https://github.com/Ray-zyy/TS-Fault"}], "authors": ["Yuyang Zhao", "Lian Xu", "Hao Miao", "Chenxi Liu", "Hao Xue"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.18539", "source_url": "https://arxiv.org/abs/2606.18539"}, "categories": ["General AI", "Robustness", "Safety & Trustworthiness", "cs.LG"], "description": {"en": "Evaluates time series forecasting models under four explicit structural fault modes (time-warped shock, dependency-fracture shock, regime-transition missingness, cascading sensor-to-system failure) injected into lookback windows, with paired clean/corrupt protocol and five difficulty levels across nine datasets and six domains."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.18539", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.18539", "title": "TS-Fault"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["The Hong Kong University of Science and Technology (Guangzhou)", "The University of Western Australia", "Hong Kong Polytechnic University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.18539", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.18539", "source_url": "https://openalex.org/W7165149536"}, "key": "claire-radar:2606.18539", "modality": null, "name": "TS-Fault", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_ts-fault_c6d5ff11", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "12e195c9b063fcf88bf13c987a6aa02e7b983a9e0272a3c4ed680643ed691786", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.18539"}, "publisher": null, "released": "2026-06-16", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.18539", "source_url": "https://arxiv.org/abs/2606.18539"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-18539", "source": "claire_radar", "source_benchmark_id": "2606.18539", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Safety & Trustworthiness", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/Ray-zyy/TS-Fault", "githubScope": "benchmark_repo", "githubStars": 9, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.18539", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "fresh"}, "capabilities": ["Robustness"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "845f0288b326849a82f6bea76f8d141fc12f501023298f99bc795ce8cf2906ca", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates time series forecasting models under four explicit structural fault modes (time-warped shock, dependency-fracture shock, regime-transition missingness, cascading sensor-to-system failure) injected into lookback windows, with paired clean/corrupt protocol and five difficulty levels across nine datasets and six domains.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "Treating TSF robustness as a data-quality problem, we present TS-Fault, a benchmark that evaluates forecasting models under explicit, parameterized fault scenarios with controllable semantic difficulty."}, "familyId": "bmf_65f522aa76b1", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/Ray-zyy/TS-Fault", "url": "https://github.com/Ray-zyy/TS-Fault"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/Ray-zyy/TS-Fault", "url": "https://huggingface.co/papers/2606.18539"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/Ray-zyy/TS-Fault.", "url": "https://arxiv.org/abs/2606.18539"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/Ray-zyy/TS-Fault", "url": "https://arxiv.org/abs/2606.18539"}], "scope": "benchmark_repo", "url": "https://github.com/Ray-zyy/TS-Fault", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_ts-fault_c6d5ff11", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/Ray-zyy/TS-Fault", "data": null, "hfPaper": "https://huggingface.co/papers/2606.18539", "paper": "https://arxiv.org/abs/2606.18539", "pdf": "https://arxiv.org/pdf/2606.18539", "project": null, "report": "https://arxiv.org/abs/2606.18539"}, "metrics": [], "motivation": "Time series forecasting (TSF) underpins consequential decisions in energy, transportation, finance, and healthcare, yet TSF models are almost universally ranked by a single number (e.g., average error) on clean held-out data, under the implicit assumption that it predicts deployed reliability.", "name": "TS-Fault", "oneLine": "Evaluates time series forecasting models under four explicit structural fault modes (time-warped shock, dependency-fracture shock, regime-transition missingness, cascading sensor-to-system failure) injected into lookback windows, with paired clean/corrupt protocol and five difficulty levels across nine datasets and six domains.", "paperTitle": "TS-Fault: Benchmarking Time Series Forecasters Against Structural Faults", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.18539", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "HKUST(GZ)", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Ray-zyy/TS-Fault"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-16", "source": {"authors": ["Yuyang Zhao", "Lian Xu", "Hao Miao", "Chenxi Liu", "and Hao Xue"], "categories": ["cs.LG", "stat.ML"], "id": "2606.18539", "title": "TS-Fault: Benchmarking Time Series Forecasters Against Structural Faults", "type": "arxiv", "url": "https://arxiv.org/abs/2606.18539"}, "sourceUpdatedAt": "2026-06-16T23:15:20Z", "topics": ["cs.LG"], "venueAttempts": [], "whyItMatters": "Standard clean-data leaderboards assume a single error metric predicts deployed reliability, but real faults are structured events. TS-Fault provides a diagnostic protocol that isolates robustness to named fault mechanisms at tunable severities, revealing that clean accuracy anti-correlates with robustness and mechanism-level faults reorder model rankings."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
