Pre-trained models have demonstrated exceptional generalization capabilities in time-series forecasting; however, adapting them to evolving data distributions remains a significant challenge. A key hurdle lies in accessing the original training data, as fine-tuning solely on new data often leads to catastrophic forgetting. To address this issue, we propose Replay Tuning (R-Tuning), a novel framework designed for the continual adaptation of pre-trained time-series models. R-Tuning constructs a unified latent space that captures both prior and current task knowledge through a frequency-aware replay strategy. Specifically, it augments model-generated samples via wavelet-based decomposition across multiple frequency bands, generating trend-preserving and fusion-enhanced variants to improve representation diversity and replay efficiency. To further reduce reliance on synthetic samples, R-Tuning introduces a latent consistency constraint that aligns new representations with the prior task space. This constraint guides joint optimization within a compact and semantically coherent latent space, ensuring robust knowledge retention and adaptation. Extensive experimental results demonstrate the superiority of R-Tuning, which reduces MAE and MSE by up to 46.9% and 46.8%, respectively, on new tasks, while preserving prior knowledge with gains of up to 5.7% and 6.0% on old tasks. Notably, under few-shot settings, R-Tuning outperforms all state-of-the-art baselines even when synthetic proxy samples account for only 5% of the new task dataset.

Introducing virtual reality technology into surgical training markedly enhances the development of surgical skills. It is critical to accurately assess surgeon proficiency post-training, as traditional evaluation methods often fail to distinguish effectively between novices and experts. Analysis of EEG signals during virtual surgical procedures, augmented by machine learning techniques, provides a robust method for improving skill assessment. The study finds that utilizing both traditional metrics and EEG rhythm indicators significantly increases classification accuracy. Specifically, when integrated with support vector machines and convolutional recurrent neural networks, these metrics enhance classification accuracy by 15.84% and 19.07%, respectively. This methodology offers a thorough tool for assessing surgical proficiency, underscoring the efficacy of EEG rhythms and machine learning in advancing surgical training evaluations.

Encoding time series into tokens and using language models for processing has been shown to substantially augment the models' ability to generalize to unseen tasks. However, existing language models for time series forecasting encounter several obstacles, including aliasing distortion and prolonged inference times, primarily due to the limitations of quantization processes and the computational demands of large models. This paper introduces Apollo-Forecast, a novel framework that tackles these challenges with two key innovations: the Anti-Aliasing Quantization Module (AAQM) and the Race Decoding (RD) technique. AAQM adeptly encodes sequences into tokens while mitigating high-frequency noise in the original signals, thus enhancing both signal fidelity and overall quantization efficiency. RD employs a draft model to enable parallel processing and results integration, which markedly accelerates the inference speed for long-term predictions, particularly in large-scale models. Extensive experiments on various real-world datasets show that Apollo-Forecast outperforms state-of-the-art methods by 35.41% and 18.99% in WQL and MASE metrics, respectively, in zero-shot scenarios. Furthermore, our method achieves an acceleration of 1.9X-2.7X in inference speed over the baseline methods.

Language models have been proven to handle time series data after tokenization and show generalization performance on unseen forecasting tasks. However, existing techniques for tokenizing time series data struggle to eliminate redundant information and noise, which can lead to signal aliasing and cumulative quantization errors, making it difficult to further improve prediction performance. In this paper, we propose an Adaptive Component Extraction and Tokenization (ACET) framework, which includes two key novelties to address these challenges: the Dynamic Component Extraction Module (DCEM) and the Time Series Tokenization Module (TSTM). The DCEM dynamically isolates the principal components from the interference in the original signal, eliminating the need for manual parameter tuning. This not only enhances the accuracy of signal tokenization but also mitigates the adverse effects of high-frequency noise. Then, the TSTM tokenizes continuous time series data while preserving long-term trend features, ensuring that critical information is retained for subsequent forecasting. Extensive cross-domain experiments on various real-world datasets demonstrate that, in zero-shot forecasting scenarios, ACET achieves improvements of 19.27% in WQL and 6.63% in MASE, compared to baseline methods.