Contamination MapTime series / Model
Time-RCD (TimeRCD)
- Organisation
- Tsinghua University (thu-sail-lab)
- Family
- Time-RCD · version 2026.04 checkpoints
- Released
- 2025-10-18 (HF repo creation date of thu-sail-lab/Time-RCD per HF API; arXiv v1 2025-09-25 (https://arxiv.org/abs/2509.21190v1); checkpoints replaced in 2026.04 per README)
- Checkpoints
- thu-sail-lab/Time-RCD (best_model/pretrain_checkpoint_best_uni.pth, best_model/pretrain_checkpoint_best_multi.pth)
- Weights
- Open · apache-2.0
- Links
- PaperModel cardCode
Training data disclosure: Full list
Level determined for 100% of benchmark cells.
- TestTest data seen0
- SeenSeen, split unclear0
- TrainTraining data seen0
- Related data seen0
- NoNot seen49
- ?Unknown0
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Other declared training data
-
custom synthetic dataset (2.5B data points) · split: unknown
Synthetic series with injected context-dependent anomalies; generation code released.
Evidence
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
arxiv.org · paper · Appendix D.5 · retrieved 2026-10-15
Benchmark cells
GIFT-Eval
Not seen Jena Weather (GIFT-Eval) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen BizITObs - Application · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen BizITObs - Service · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen BizITObs - L2C · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Bitbrains - Fast Storage · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Bitbrains - rnd · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Restaurant (Recruit Restaurant Visitor Forecasting) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen ETT1 (GIFT-Eval) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen ETT2 (GIFT-Eval) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Loop Seattle (LibCity) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen SZ-Taxi (LibCity) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen M_DENSE (LibCity) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Solar (LSTNet, 137 Alabama PV plants) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Hierarchical Sales (Mancuso et al., 2020) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen M4 Yearly · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen M4 Quarterly · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen M4 Monthly · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen M4 Weekly · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen M4 Daily · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen M4 Hourly · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Hospital · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen COVID Deaths · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen US Births · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Saugeen River Flow · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Temperature Rain · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen KDD Cup 2018 · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Car Parts · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen ElectricityLoadDiagrams20112014 (UCI) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Long-horizon forecasting (ETT, Electricity, Traffic, Weather, Exchange, ILI)
Not seen ETTh1 · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen ETTh2 · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen ETTm1 · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen ETTm2 · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Electricity (ECL, 321 clients) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Traffic (PeMS San Francisco Bay, 862 sensors) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Weather (Jena, 2020; Autoformer) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Exchange-Rate · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen ILI (CDC influenza-like illness, national) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Time-series anomaly detection
Not seen TSB-AD-U (TSB-AD univariate) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen TSB-AD-M (TSB-AD multivariate) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen UCR Time Series Anomaly Archive (Hexagon ML/UCR, KDD Cup 2021) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Server Machine Dataset (SMD) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Mars Science Laboratory (MSL) telemetry anomaly dataset · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Soil Moisture Active Passive (SMAP) telemetry anomaly dataset · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Pooled Server Metrics (PSM) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Secure Water Treatment (SWaT) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Yahoo S5 Labeled Anomaly Detection Dataset (Yahoo Webscope S5) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen Numenta Anomaly Benchmark (NAB) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Time-series classification
Not seen UCR Time Series Classification Archive (2018) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Not seen UEA Multivariate Time Series Classification Archive (2018) · absent from the complete list of training data
Evidence (2)
“Pre-training Dataset Details We trained our model only on our custom synthetic dataset, which comprises a total of 2.5B data points.”
“We also release the [pre-trained dataset generation code and hyperparameters](https://github.com/thu-sail-lab/TSAD_dataset_gen_public).”
Notes
Zero-shot anomaly FM pre-trained only on synthetic data (also in arXiv v1). Univariate and multivariate checkpoints. README: '2026.04: With a new dataset and new checkpoints' - the current checkpoints differ from the 2025 ones; the v5 paper (May 2026) still describes synthetic-only data. The paper excludes DADA and MOMENT baseline results because their pre-training overlaps its TSB-AD-based test sets.
Sources
- Towards Foundation Models for Zero-Shot TSAD: TimeRCD (arXiv v5, ICML 2026) · paper · retrieved 2026-10-15
- thu-sail-lab/Time-RCD README · readme · retrieved 2026-10-15
- thu-sail-lab/Time-RCD model card · model card · retrieved 2026-10-15