MLP-Mixer based Masked Autoencoders Are Effective, Explainable and Robust for Time Series Anomaly Detection

Summary: MMA: MLP-Mixer backbone + masked autoencoder enabling 10–20× larger input windows to detect long-duration subsequence anomalies, with contrastive learning to catch subtle anomalies. Dynamic anomaly filtering reduces false positives; robust to training contamination and provides explainable normal-pattern reconstructions. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 14425
Venue: VLDB
Year: 2025
Pagerank: 5.093636e-05
Overall Rank: 11,099 | 23.86%
DOI: 10.14778/3712221.3712243

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Qideng Tang (National University of Defense Technology)
2. Chaofan Dai (National University of Defense Technology)
3. Yahui Wu (National University of Defense Technology)
4. Haohao Zhou (National University of Defense Technology)

BibTeX Citation

@article{tang_vldb25,
        title = {{MLP-Mixer based Masked Autoencoders Are Effective, Explainable and Robust for Time Series Anomaly Detection}},
        author = {Tang, Qideng and Dai, Chaofan and Wu, Yahui and Zhou, Haohao},
        journal = {PVLDB},
        series = {{VLDB} '25},
        volume = {18},
        number = {3},
        pages = {798--812},
        doi = {10.14778/3712221.3712243},
        url = {https://doi.org/10.14778/3712221.3712243},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 0 of 0 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank

Outgoing Citations (Sorted by Pagerank)

Showing 9 of 9 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
1,029	Anomaly Detection in Time Series: A Comprehensive Evaluation	2022	VLDB	0.00012557065
1,534	Exathlon: A Benchmark for Explainable Anomaly Detection over Time Series	2021	VLDB	0.00010464308
1,793	TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data	2022	VLDB	9.7435472e-05
1,805	SAND: Streaming Subsequence Anomaly Detection	2021	VLDB	9.7116108e-05
2,004	TSB-UAD: An End-to-End Benchmark Suite for Univariate Time-Series Anomaly Detection	2022	VLDB	9.3207067e-05
2,346	Series2Graph: Graph-based Subsequence Anomaly Detection for Time Series	2020	VLDB	8.7168932e-05
3,447	Volume Under the Surface: A New Accuracy Evaluation Measure for Time-Series Anomaly Detection	2022	VLDB	7.4085215e-05
3,987	Choose Wisely: An Extensive Evaluation of Model Selection for Anomaly Detection in Time Series	2023	VLDB	6.9722766e-05
5,420	ImDiffusion: Imputed Diffusion Models for Multivariate Time Series Anomaly Detection	2024	VLDB	6.2246363e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	5,986	AutoTSAD: Unsupervised Holistic Anomaly Detection for Time Series Data	2024	VLDB
2	10,977	TSB-AutoAD: Towards Automated Solutions for Time-Series Anomaly Detection	2025	VLDB
3	4,851	Unsupervised Time Series Outlier Detection with Diversity-Driven Convolutional Ensembles	2022	VLDB
4	8,229	Mining Approximate Top-K Subspace Anomalies in Multi-Dimensional Time-Series Data	2007	VLDB
5	1,793	TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data	2022	VLDB
6	11,300	Time-Series Anomaly Detection: Overview and New Trends	2024	VLDB
7	3,987	Choose Wisely: An Extensive Evaluation of Model Selection for Anomaly Detection in Time Series	2023	VLDB
8	8,915	A Multi-Scale Decomposition MLP-Mixer for Time Series Analysis	2024	VLDB
9	6,141	An Experimental Evaluation of Anomaly Detection in Time Series	2024	VLDB
10	1,029	Anomaly Detection in Time Series: A Comprehensive Evaluation	2022	VLDB