Pith. sign in

REVIEW 17 cited by

Anomaly Transformer: Time Series Anomaly Detection with Association Discrepancy

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2110.02642 v5 pith:K4ON45AV submitted 2021-10-06 cs.LG

Jiehui Xu , Haixu Wu , Jianmin Wang , Mingsheng Long This is my paper · ORCID
classification cs.LG
keywords associationanomalyseriestimediscrepancypointsdetectionemph
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Unsupervised detection of anomaly points in time series is a challenging problem, which requires the model to derive a distinguishable criterion. Previous methods tackle the problem mainly through learning pointwise representation or pairwise association, however, neither is sufficient to reason about the intricate dynamics. Recently, Transformers have shown great power in unified modeling of pointwise representation and pairwise association, and we find that the self-attention weight distribution of each time point can embody rich association with the whole series. Our key observation is that due to the rarity of anomalies, it is extremely difficult to build nontrivial associations from abnormal points to the whole series, thereby, the anomalies' associations shall mainly concentrate on their adjacent time points. This adjacent-concentration bias implies an association-based criterion inherently distinguishable between normal and abnormal points, which we highlight through the \emph{Association Discrepancy}. Technically, we propose the \emph{Anomaly Transformer} with a new \emph{Anomaly-Attention} mechanism to compute the association discrepancy. A minimax strategy is devised to amplify the normal-abnormal distinguishability of the association discrepancy. The Anomaly Transformer achieves state-of-the-art results on six unsupervised time series anomaly detection benchmarks of three applications: service monitoring, space & earth exploration, and water treatment.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. AnomSeer: Reinforcing Multimodal LLMs to Reason for Time-Series Anomaly Detection

    cs.LG 2026-02 conditional novelty 7.0 of 10

    New RL approach (TimerPO) with ground-truth-generated expert reasoning traces lets 3B-7B multimodal LLMs outperform GPT-4o on time-series anomaly detection and explanation.

  2. Argos: Agentic Time-Series Anomaly Detection with Autonomous Rule Generation via Large Language Models

    cs.LG 2025-01 reject novelty 7.0 of 10

    ARGOS uses LLM agents to generate explainable, reproducible anomaly detection rules and fuses them with a base detector, reporting higher F1 than deep-learning and LLM baselines on KPI, Yahoo, and a Microsoft internal...

  3. Onnes: A Physics-Grounded Multi-Agent LLM Simulator for Cryogenic Fault Diagnosis in Quantum Computing Infrastructure

    cs.AI 2026-07 conditional novelty 6.0 of 10

    A physics-grounded dilution-fridge simulator with LLM agents achieves supervised-ML parity on cryogenic fault classification using six demonstrations and no training, validated on simulated telemetry plus a real-hardw...

  4. Surprised by Attention: Predictable Query Dynamics for Time Series Anomaly Detection

    cs.LG 2026-03 conditional novelty 6.0 of 10

    Predicting multi-head attention queries from history and scoring cosine mismatch against an EMA target, combined with reconstruction error, improves unsupervised multivariate anomaly ranking and localization.

  5. Cluster-Aware Causal Mixer for Online Anomaly Detection in Multivariate Time Series

    cs.LG 2025-05 conditional novelty 6.0 of 10

    A cluster-aware causal MLP-mixer with sequential evidence-accumulation scoring achieves state-of-the-art best-F1 on six multivariate time-series anomaly detection benchmarks.

  6. Federated Foundation Models on Heterogeneous Time Series

    cs.LG 2024-12 conditional novelty 6.0 of 10

    FFTS is a federated pretraining framework with a timescale-aware mixture-of-experts module that trains a time series foundation model from scratch across heterogeneous, non-shared datasets.

  7. TSINR: Capturing Temporal Continuity via Implicit Neural Representations for Time Series Anomaly Detection

    cs.LG 2024-11 conditional novelty 6.0 of 10

    TSINR detects anomalies by reconstructing time series with a transformer-predicted implicit neural representation that favors low-frequency signals.

  8. Benchmarking Unsupervised Strategies for Anomaly Detection in Multivariate Time Series

    cs.LG 2025-06 conditional novelty 5.0 of 10

    Across ten public datasets, a reconstruction-based inverted transformer with per-variate anomaly labelling achieves the best or tied best MCC on most datasets, but the comparison is weakened by test-set-based configur...

  9. Multivariate Time Series Anomaly Detection by Capturing Coarse-Grained Intra- and Inter-Variate Dependencies

    cs.LG 2025-01 conditional novelty 5.0 of 10

    MtsCID combines frequency-domain transformers, multi-scale patch attention, and fixed sinusoidal prototypes to detect anomalies, outperforming nine baselines on six of seven datasets.

  10. Noise-Resilient Point-wise Anomaly Detection in Time Series Using Weak Segment Labels

    cs.LG 2025-01 conditional novelty 5.0 of 10

    NRdetector combines positive-unlabeled learning, confidence-based sample selection, and a temporal smoothness-and-separability loss to predict point-level anomalies from noisy segment labels, and reports the best F1 o...

  11. A New Perspective on Time Series Anomaly Detection: Faster Patch-based Broad Learning System

    cs.LG 2024-12 conditional novelty 5.0 of 10

    A shallow patch-based broad learning system with a random-perturbation contrastive branch and multi-scale patch ensembling reports state-of-the-art unsupervised time series anomaly detection on five benchmarks.

  12. Attention in Motion: Secure Platooning via Transformer-based Misbehavior Detection

    cs.CR 2025-12 conditional novelty 4.0 of 10

    A transformer with global positional offsets and a false-positive-penalizing loss detects kinematic falsification attacks in simulated platoons at 0.13-0.8 ms inference.

  13. Rethinking Spatio-Temporal Anomaly Detection: A Vision for Causality-Driven Cybersecurity

    cs.LG 2025-07 unverdicted novelty 4.0 of 10

    The paper is a position piece advocating causal graph learning as the basis for interpretable, drift-robust anomaly detection in cyber-physical systems, with a small comparison table as supporting evidence.

  14. Exploring a Hybrid Deep Learning Approach for Anomaly Detection in Mental Healthcare Provider Billing: Addressing Label Scarcity through Semi-Supervised Anomaly Detection

    cs.LG 2025-07 conditional novelty 4.0 of 10

    Pseudo-labeling with Isolation Forest lets an LSTM detect billing anomalies with high recall, but the hybrid LSTM-Transformer adds little and reduces precision.

  15. Temporal cross-validation impacts multivariate time series subsequence anomaly detection evaluation

    stat.ML 2025-06 conditional novelty 4.0 of 10

    Sliding-window cross-validation yields higher and more stable AUC-PR than walk-forward for subsequence anomaly detection on two small CAN fault datasets.

  16. Graph Neural Network and Transformer Integration for Unsupervised System Anomaly Discovery

    cs.LG 2025-08 reject novelty 3.0 of 10

    GTF-Net, a GCN-Transformer fusion, reports F1 0.889 and AUC 0.942 for anomaly detection on Alibaba cluster data, but omits the training objective and code.

  17. Predictive Modeling, Pattern Recognition, and Spatiotemporal Representations of Plant Growth in Simulated and Controlled Environments: A Comprehensive Review

    q-bio.QM 2024-12 conditional novelty 3.0 of 10

    A review of 51 papers and 22 datasets finds plant growth modeling moving from static regressions to probabilistic, image-generating models with environmental feedback.

Pith tools