arXiv:2606. 05138v1 Announce Type: new Abstract: Generating realistic financial time series is challenging as training data is often limited to a single historical path.
arXiv:2607. 15774v1 Announce Type: cross Abstract: Explainable AI (XAI) for time series has seen significant algorithmic growth, but its utility in providing measurable performance gains for downstream tasks remains under-explored.
arXiv:2511. 02152v2 Announce Type: replace Abstract: Time series data is one of the most popular data modalities in critical domains such as industry and medicine.
KiT is a K‑line Diffusion Transformer foundation model designed for financial time‑series forecasting. It reframes future prediction as conditional path generation via flow matching, producing ensembles of plausible OHLCV trajectories from a historical context window. Trained on billions of candlestick bars across multiple markets and timescales, KiT achieves superior RankIC scores compared to task‑specific forecasters and general time‑series models.
arXiv:2605.01418v2 Announce Type: replace Abstract: Time-series data are inherently multiscale, spanning diverse temporal granularities from coarse trends to fine-scale dynamics. However, existing ti...
The paper proposes a hybrid generative adversarial network (GAN) that combines a bi-directional LSTM and a CNN (Bi‑LSTM‑CNN) to generate synthetic financial data aligned with real market data. By preserving stock trend features, the model predicts future stock price movements across multiple markets (TSX, SHCOMP, S&P 500). Experiments show that this hybrid approach outperforms existing machine‑learning prototypes, and the study highlights gaps between investors and technical researchers.
arXiv:2608. 09193v1 Announce Type: cross Abstract: Unsupervised time-series domain adaptation (DA) addresses the challenge of transferring a classifier from a labeled source domain to an unlabeled target domain under distribution shifts induced by different users, sensors, devices, acquisition conditions, or temporal dynamics.
arXiv:2602. 16224v2 Announce Type: replace Abstract: Time series data are prone to noise in various domains, and training samples may contain low-predictability patterns that deviate from the normal data distribution, leading to training instability or convergence to poor local minima.
arXiv:2608. 06223v1 Announce Type: new Abstract: While deep learning models, particularly transformer-based architectures, have shown impressive performance in time series forecasting, the application of retrieval-augmented generation (RAG) in this domain remains limited.
arXiv:2606. 15172v1 Announce Type: new Abstract: Synthesizing realistic time series with generative models has wide-ranging applications in real-world scenarios.
The paper introduces a convolutional neural network framework for detecting event-driven dynamics in univariate time‑series windows, showing that it can represent classifiers based on range, maximum drawup, maximum drawdown, and slope change, and can uniformly approximate realised volatility and autoregressive explosiveness. It provides error bounds for representative rules in finite samples and an oracle inequality for learning across them, with simulations indicating that the model matches or outperforms individual‑statistic classifiers as training data increases. In an application to six daily energy price series, a hierarchical CNN identifies event windows and families, correctly detecting geopolitical dynamics around the 2026 Iran war and a natural gas spike linked to weather without retraining on post‑February 2026 data.
arXiv:2410. 07299v3 Announce Type: replace-cross Abstract: We introduce OTIS, an open time series encoder that yields high-quality time series features for downstream deployment on any system, including resource-constrained wearables and industrial sensors.
arXiv:2607. 07500v1 Announce Type: cross Abstract: Time series classification (TSC) is dominated by a two-stage paradigm: train a feature encoder -- either from scratch on the target dataset or via pretraining on large corpora -- and then fit a task-specific classifier on top.