pith. sign in

Recoverable Identifier

arXiv:2605.05725 · detector doi_compliance · incontrovertible · 2026-05-19 13:18:31.063764+00:00

advisory doi_compliance recoverable_identifier

DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.48550/arXiv.2411.02465.13) was visible in the surrounding text but could not be confirmed against doi.org as printed.

Paper page Integrity report arXiv Try DOI

Evidence text

Jiaxin Zhuang, Leon Yan, Zhenwei Zhang, Ruiqi Wang, Jiawei Zhang, and Yuantao Gu. See it, think it, sorted: Large multimodal models are few-shot time series anomaly analyzers.CoRR, abs/2411.02465, 2024. doi: 10.48550/ARXIV .2411.02465. URL https://doi.org/10. 48550/arXiv.2411.02465. 13 Table 4: Definitions and synthetic injection rules of the nine anomaly types used in SAGE. ID Type Name Family Definition Synthetic Injection Rule 1 Global point anomaly Point A point that strongly deviates from the global distribution of the series Insert a large isolated spike or drop at a random position. 2 Contextual point anomaly Point A point that appears abnormal only under its local temporal context Inject a local spike or drop within an otherwise smooth region so that it is inconsistent with nearby values. 3 Amplitude change Seasonal A change in the magnitude of a recurring or oscillatory pattern Multiply a periodic segment by an amplitude scaling factor while preserving its frequency. 4 Seasonality anomaly Seasonal A disruption or change in periodic structure or dominant frequency Alter the period or frequency of a periodic segment, or partially break its regular repetition. 5 Trend change Structural A persistent change in the slope or long-term direction of the series Add a piecewise linear drift or slope change after a selected point. 6 Mean change point Structural An abrupt shift in the mean level of the series Add a step-like level shift after a selected change point. 7 Variance c

Evidence payload

{
  "printed_excerpt": "Jiaxin Zhuang, Leon Yan, Zhenwei Zhang, Ruiqi Wang, Jiawei Zhang, and Yuantao Gu. See it, think it, sorted: Large multimodal models are few-shot time series anomaly analyzers.CoRR, abs/2411.02465, 2024. doi: 10.48550/ARXIV .2411.02465. URL ",
  "reconstructed_doi": "10.48550/arXiv.2411.02465.13",
  "ref_index": 40,
  "resolved_title": null,
  "verdict_class": "incontrovertible"
}