AI Paper Atlas

Research map / Time series / Anomaly detection

Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancy

Tian Lan, Hao Duong Le, Jinbo Li, Wenjun He, Meng Wang, Chenghao Liu, Chen Zhang

ICML 2026 · Time series › Anomaly detection · approach: Time series anomaly detection evaluation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use anomaly head instead of reconstruction-error inference (ablation)TimeRCD · UCRVUS-PR: 4.08 -> 33.57Improved
add anomaly head supervision to reconstruction-only training (ablation)TimeRCD · UCRVUS-PR: 3.35 -> 33.57Improved
add masked-reconstruction auxiliary loss to anomaly-only training (ablation)TimeRCD · UCRVUS-PR: 5.90 -> 33.57Improved
jointly train encoder instead of frozen reconstruction-pretrained encoder (ablation)TimeRCD · UCRVUS-PR: 26.11 -> 33.57Improved
use synthetic data with in-context anomaly injection instead of real data with DADA injection (ablation)TimeRCD · univariate benchmarksF1-T: 0.073 -> 0.569Improved
use in-context injection instead of DADA injection on synthetic series (ablation)TimeRCD · univariate benchmarksF1-T: 0.505 -> 0.569Improved
use in-context injection instead of DADA injection on synthetic series (ablation)TimeRCD · univariate benchmarksVUS-PR: 0.478 vs 0.487No clear effect
scale synthetic pretraining data from 350M to 2.5B points (ablation)TimeRCD · avg VUS-PR across benchmarksVUS-PRImproved
increase context window from 1k to 13k (ablation)TimeRCD · UCR, Power, SMAP, SMDdetection performanceImproved
increase context window (ablation)TimeRCD · YAHOOdetection performanceNo clear effect
use TimeRCD instead of foundation baselines (zero-shot)TimeRCD · 16 datasets x 4 metricsrank: first in 38 and second in 14 of 64 casesImproved
use large context for contextual anomaliesTimeRCD · synthetic contextual-anomaly splitStandard-F1: 0.827 vs competitors degradeImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy