AI Paper Atlas

Research map / Benchmarks / SMAP

SMAP leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on SMAP. Best F1: 96.72 (PUAD, ICML 2023).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

SMAP · F1

10 papers, each with its best reported number for its own method; higher is better.

#F1 ↑MethodPaperTopic
196.72PUADPrototype-oriented unsupervised anomaly detection for multivariate time series
ICML 2023 · Table 1
Anomaly detection
296.69Anomaly TransformerAnomaly Transformer: Time Series Anomaly Detection with Association Discrepancy
ICLR 2022 · Table: Quantitative results
Anomaly detection
396.61MEMTO
Transformer
MEMTO: Memory-guided Transformer for Multivariate Time Series Anomaly Detection
NeurIPS 2023 · Table 1
Anomaly detection
491.4DVGCRN-layer3-MDeep Variational Graph Convolutional Recurrent Network for Multivariate Time Series Anomaly Detection
ICML 2022 · Table 1
Anomaly detection
575.42DADA (zero shot)Towards a General Time Series Anomaly Detector with Adaptive Bottlenecks and Dual Adversarial Decoders
ICLR 2025 · Table Results for five real-wo
Anomaly detection
670.74FITSFITS: Modeling Time Series with $10k$ Parameters
ICLR 2024 · Anomaly detection result of F1
Anomaly detection
751.1NPSR (combined)
Performer
Nominality Score Conditioned Time Series Anomaly Detection by Point/Sequential Reconstruction
NeurIPS 2023 · Table 2
Anomaly detection
848ChInf + GCN-LSTM
1-Layer GCN-LSTM
Channel Matters: Estimating Channel Influence for Multivariate Time Series
NeurIPS 2025 · Table: anomaly detection resul
Anomaly detection
934TimeInf
Linear AR
TimeInf: Time Series Data Contribution via Influence Functions
ICLR 2025 · Table: Anomaly detection perfo
Anomaly detection
1028.01ICDiffADICDiffAD: Implicit Conditioning Diffusion Model for Time Series Anomaly Detection
ICLR 2026 · Table 1
Anomaly detection

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.