Time series and spatio-temporal data

CondTSC — Dataset Condensation for Time Series Classification via Dual Domain Matching

Zhanyu Liu, Ke Hao, Guanjie Zheng et al. · KDD 2024Time seriespaper ↗code ↗

CondTSC adapts image-style dataset condensation to time series classification with dual surrogate matching objectives computed in both the time and frequency domains (rather than time domain alone), combined with multi-view data augmentation and dual-domain training, to better capture the temporal and spectral structure that direct adaptations of image condensation methods miss. The paper reports outperforming other time-series and adapted image/graph condensation baselines, producing a condensed set that better matches the real data's distribution.

TimeDC — Less is More: Efficient Time Series Dataset Condensation via Two-fold Modal Matching--Extended Version

Hao Miao, Ziqiao Liu, Yan Zhao et al. · VLDB 2025Time seriespaper ↗code ↗

TimeDC condenses time series with two-fold modal matching, decomposition-driven frequency matching to preserve spectral structure and curriculum training-trajectory matching (with a buffer of precomputed expert trajectories) to preserve temporal dependencies, aiming for effective and generalized condensation rather than a single-domain surrogate objective. The paper reports effectiveness and efficiency gains over prior time-series condensation baselines across real datasets.

CondTSC — Dataset Condensation for Time Series Classification via Dual Domain Matching

Zhanyu Liu, Ke Hao, Guanjie Zheng et al. · KDD 2024Time seriespaper ↗code ↗

CondTSC adapts image-style dataset condensation to time series classification with dual surrogate matching objectives computed in both the time and frequency domains (rather than time domain alone), combined with multi-view data augmentation and dual-domain training, to better capture the temporal and spectral structure that direct adaptations of image condensation methods miss. The paper reports outperforming other time-series and adapted image/graph condensation baselines, producing a condensed set that better matches the real data's distribution.

TimeDC — Less is More: Efficient Time Series Dataset Condensation via Two-fold Modal Matching--Extended Version

Hao Miao, Ziqiao Liu, Yan Zhao et al. · VLDB 2025Time seriespaper ↗code ↗

TimeDC condenses time series with two-fold modal matching, decomposition-driven frequency matching to preserve spectral structure and curriculum training-trajectory matching (with a buffer of precomputed expert trajectories) to preserve temporal dependencies, aiming for effective and generalized condensation rather than a single-domain surrogate objective. The paper reports effectiveness and efficiency gains over prior time-series condensation baselines across real datasets.

UniTSC — One Batch Is Enough: A Unified Dataset Condensation Framework for General Time Series Analysis

Wei Shao, Ziquan Fang, Zheqi Lu et al. · ICML 2026Time seriespaper ↗code ↗

UniTSC argues that prior time-series condensation methods are task-specific (optimized for one downstream task such as forecasting and performing poorly on others such as imputation), and proposes a task-invariant condensation framework that jointly captures temporal, spectral and topological properties of the data so the same condensed set generalizes across multiple time-series analysis tasks.

ShapeCond — ShapeCond: Fast Shapelet-Guided Dataset Condensation for Time Series Classification

Sijia Peng, Yun Xiong, Xi Chen et al. · arXiv 2026Time seriespaper ↗code ↗

ShapeCond condenses time series for classification by extracting shapelets, discriminative local motifs that image-centric condensation methods miss, and using a shapelet-guided optimization strategy to preserve them explicitly, with a synthesis cost that is independent of sequence length. The paper reports large speedups over the prior state of the art (CondTSC), e.g. 29x faster and up to 10,000x faster than naively using shapelets on a 3,000-timestep dataset, while improving downstream accuracy over prior time-series condensation methods.

Harmonic Dataset Distillation for Time Series Forecasting

Seungha Hong, Sanghwan Jang, Wonbin Kweon et al. · AAAI 2026notableTime seriespaper ↗

HDT (Harmonic Dataset Distillation) decomposes each time series into its sinusoidal basis via the FFT and condenses by Harmonic Matching, aligning the core periodic structure of the synthetic and real series in the frequency domain so that updates never disturb temporal dependencies. The paper reports strong cross-architecture generalization and scalability for time series forecasting compared to conventional, image-centric condensation methods adapted to time series.

CondTSF — CondTSF: One-line Plugin of Dataset Condensation for Time Series Forecasting

Jianrong Ding, Zhanyu Liu, Guanjie Zheng et al. · NeurIPS 2024Time seriespaper ↗code ↗

Analyzes why classification-style dataset-condensation objectives are misaligned with time-series forecasting, where evaluation requires closeness of predictions rather than matching discrete labels, and proposes CondTSF, a one-line plugin that can be added to existing time-series condensation methods to shrink the gap between full-data and synthetic-data forecast predictions; reports consistent gains across eight benchmark datasets, largest at low condensation ratios.

DDTime — DDTime: Dataset Distillation with Spectral Alignment and Information Bottleneck for Time-Series Forecasting

Yuqi Li, Kuiye Ding, Chuanguang Yang et al. · arXiv 2025Time seriespaper ↗

Proposes DDTime, a lightweight plug-in for time-series-forecasting dataset distillation that corrects autocorrelation-induced value-alignment bias with a frequency-domain alignment mechanism and increases synthetic-sample diversity with an information-bottleneck-inspired inter-sample regularizer; reports about 30% relative accuracy gains over prior distillation methods across 20 benchmark datasets and several forecasting architectures, at roughly 2.49% computational overhead.

STemDist — Effective Dataset Distillation for Spatio-Temporal Forecasting with Bi-dimensional Compression

Taehyung Kwon, Yeonje Choi, Yeongho Kim et al. · ICDE 2026Time seriespaper ↗code ↗

Proposes STemDist, the first dataset-distillation method for spatio-temporal forecasting that compresses both the temporal and spatial dimensions in a balanced way, performing coarse-grained distillation at the level of location clusters and complementing it with a subset-based granular technique within each cluster; reports up to 6x faster training, 8x lower memory use, and up to 12% lower prediction error than prior general and time-series distillation methods across five real-world datasets.