Anomaly detection: research map
86 papers on Anomaly detection in Time series, from ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026), grouped into 1 clusters and 2 approaches. The busiest year so far is 2026.
What it is
Finding unusual events in time series, such as sensor faults, attacks or fraud, often across many variables at once, and evaluating such detectors fairly.
Terms you will see on this page:
- Time series anomaly
- a point or segment that departs from normal behaviour.
- Multivariate
- many signals recorded together.
- Point-adjust evaluation
- a common but criticised scoring rule for time-series detectors.
Start reading with the most cited papers here:
- Graph Neural Network-Based Anomaly Detection in Multivariate Time Series (AAAI 2021 · 1,301 citations)
- Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy (ICLR 2022 · 242 citations)
Within Time series, its share grew from 3.7% in 2023–24 to 7.5% in 2025–26 (14 → 59 papers at ICML, NeurIPS, ICLR, CVPR and AAAI, the venues with data for all four years).
Explore Anomaly detection in the interactive map
Working on something in this topic? Describe your idea in scime atlas to see which approach it falls under, the closest papers by meaning and how crowded the spot has become.
What changed recently
Approaches whose share of the topic moved most between 2023–24 and 2025–26 (ICML, NeurIPS, ICLR, CVPR and AAAI):
- Multivariate anomaly detection rising: 43% → 46% of the topic (6 → 27 papers)
Papers from 2025–2026 that are already cited most:
- Graph Mixture of Experts and Memory-augmented Routers for Multivariate Time Series Anomaly Detection (AAAI 2025 · 22 citations)
- GCAD: Anomaly Detection in Multivariate Time Series from the Perspective of Granger Causality (AAAI 2025 · 18 citations)
Approaches and key papers
Time series anomaly detection · 86 papers
Keywords: anomaly, detection, anomalies
Approaches in this cluster:
- Time series anomaly detection evaluation (48 papers)
Evaluate and improve anomaly detection on time series with metrics, foundation models and LLMs. - Multivariate anomaly detection (38 papers)
Detect anomalies across series using graph dynamics, causality and diffusion reconstruction.
Most cited and most cited since 2024:
- Graph Neural Network-Based Anomaly Detection in Multivariate Time Series (AAAI 2021 · 1,301 citations)
- Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy (ICLR 2022 · 242 citations)
- The Elephant in the Room: Towards A Reliable Time-Series Anomaly Detection Benchmark (NeurIPS 2024 · 51 citations)
- When Model Meets New Normals: Test-Time Adaptation for Unsupervised Time-Series Anomaly Detection (AAAI 2024 · 34 citations)
