# Temporal data

Data whose values, availability, or validity depend on time, commonly represented as time series, event sequences, or timestamped records.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## TRESS: A Fixed-Size History Representation for Irregular Time Series Forecasting

DevFeed: [TRESS: A Fixed-Size History Representation for Irregular Time Series Forecasting](<https://devfeed.tech/articles/tress-a-fixed-size-history-representation-for-irregular-time-series-forecasting-143500.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.24257>)

Author: Shusen Ma, Yangchun Xie, Xudong Zuo, Haizhou Liang, Yun-Bo Zhao

Published: 2026-11-19T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Time-sequences](<https://devfeed.tech/topics/time-sequences.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [cpu-inference](<https://devfeed.tech/tags/cpu-inference.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [fixed](<https://devfeed.tech/tags/fixed.md>), [history](<https://devfeed.tech/tags/history.md>), [size](<https://devfeed.tech/tags/size.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [time-series-forecasting](<https://devfeed.tech/tags/time-series-forecasting.md>)

### AI overview

TRESS represents irregular time series with a fixed-size history: it keeps each variable's latest observations at their original timestamps and summarizes the full history, so representation size stays constant as records grow. Its query-conditioned event lifting weights deviations from the latest value by observation age and forecast horizon; a channel-specific decoder combines these features with historical statistics and a cross-variable summary to predict a correction. With 8-16 retained observations per variable, TRESS achieved the lowest mean test MSE and MAE among methods compared on all four benchmarks. Reusing one encoded history for 32 successive query times reduced CPU inference time by 2.12-2.60x across those datasets.

### Source excerpt

arXiv:2611.24257v1 Announce Type: new Abstract: A fixed number of recent observations may span minutes for one variable and days for another. Earlier measurements can still reveal whether the latest value is unusual for that variable. We propose TRESS (Timestamped Recent Events and Statistical Summaries), a forecasting model that retains the latest observations at their original times and summarises the full history of each variable. For a given set of variables and event budget, this representation stays the same size as the record grows. Its query-conditioned event lifting module weights deviations from the latest value according to the age of each observation and the forecast horizon. A channel-specific decoder combines the resulting features with historical statistics and a summary shared across variables to predict a correction to the latest value. Retaining 8 to 16 observations per variable, TRESS records the lowest mean test MSE and MAE among the methods compared on all four benchmarks. Serving 32 successive query times from one encoded history reduces CPU inference time by a factor of 2.12--2.60 across the four datasets. The code is at https://github.com/SSMa913/NICLab-TRESS.

## Reason over time-series data with OpenTSLM on Amazon SageMaker JumpStart

DevFeed: [Reason over time-series data with OpenTSLM on Amazon SageMaker JumpStart](<https://devfeed.tech/articles/reason-over-time-series-data-with-opentslm-on-amazon-sagemaker-jumpstart-138239.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/reason-over-time-series-data-with-opentslm-on-amazon-sagemaker-jumpstart/>)

Author: Martin Maritsch

Published: 2026-11-17T18:06:44Z

Content type: article

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [Amazon SageMaker JumpStart](<https://devfeed.tech/topics/amazon-sagemaker-jumpstart.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Time Series](<https://devfeed.tech/topics/time-series.md>), [Amazon SageMaker](<https://devfeed.tech/topics/amazon-sagemaker.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [amazon-sagemaker](<https://devfeed.tech/tags/amazon-sagemaker.md>), [amazon-sagemaker-jumpstart](<https://devfeed.tech/tags/amazon-sagemaker-jumpstart.md>), [deploy](<https://devfeed.tech/tags/deploy.md>), [it-support](<https://devfeed.tech/tags/it-support.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [proactive](<https://devfeed.tech/tags/proactive.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [technical-how-to](<https://devfeed.tech/tags/technical-how-to.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [time-series-data](<https://devfeed.tech/tags/time-series-data.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

OpenTSLM-Flamingo, a 5.24-billion-parameter model combining Gemma 3 4B with a time-series encoder and cross-attention, is available through SageMaker JumpStart for deployment to a real-time endpoint in an AWS environment. Requests pair numerical series with questions and optional context; each series can contain up to 32,768 values. A request supports up to four instances and 32 series per instance, but a 60-second invocation limit and sequential processing constrain long answers or batches. The text also recommends testing input windows because evaluated time-series models can lose event-retrieval accuracy on longer histories. For adaptation, a SageMaker training job loads ARFBench through TimeNet, fine-tunes selected model components while freezing the rest of Gemma 3, and stores the resulting artifact in S3. In the reported run, held-out ARFBench accuracy increased from 5.0% to 45.9%; the article says this is an example result, not a model comparison or official leaderboard score. It recommends evaluating representative cases before deciding whether to improve context or fine-tune.

### Source excerpt

OpenTSLM, an open source family of time-series language models, is now available in Amazon SageMaker JumpStart. See how Nexthink and Aionic Labs apply it to proactive IT support, then deploy OpenTSLM to a real-time endpoint and fine-tune it on your own data with TimeNet and a SageMaker training job.

## Look Back, Correct Forward: Error Feedback Calibration for Time Series Forecasting

DevFeed: [Look Back, Correct Forward: Error Feedback Calibration for Time Series Forecasting](<https://devfeed.tech/articles/look-back-correct-forward-error-feedback-calibration-for-time-series-forecasting-137295.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.22872>)

Author: Haonan Yang, Yulin He, Zhuo Li, Jing Zhao, Huibin Tan

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Predictive AI](<https://devfeed.tech/topics/predictive-ai.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [controls](<https://devfeed.tech/tags/controls.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [error](<https://devfeed.tech/tags/error.md>), [errors](<https://devfeed.tech/tags/errors.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [forecast](<https://devfeed.tech/tags/forecast.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [framework](<https://devfeed.tech/tags/framework.md>), [future](<https://devfeed.tech/tags/future.md>), [learning](<https://devfeed.tech/tags/learning.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [out-of-sample](<https://devfeed.tech/tags/out-of-sample.md>), [performance](<https://devfeed.tech/tags/performance.md>), [post](<https://devfeed.tech/tags/post.md>), [predictions](<https://devfeed.tech/tags/predictions.md>), [series](<https://devfeed.tech/tags/series.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [time-series-forecasting](<https://devfeed.tech/tags/time-series-forecasting.md>)

### AI overview

Error Feedback Calibration (EFC) uses revealed errors from overlapping past forecasts to correct future predictions while leaving the forecasting model frozen. It builds a Causal Error Diagonal (CED) to represent recent failure behavior, then uses Cumulative Error Transport (CET) to estimate future residuals; temporal out-of-sample validation controls the correction strength. Temporal Residual Learning (TRL) predicts only the residual structure left after CET, using recent CED trajectories. The abstract reports improvements across forecasting backbones and datasets while substantially controlling harmful overcorrection; it does not provide specific results or implementation details.

### Source excerpt

arXiv:2610.22872v2 Announce Type: replace Abstract: Time series forecasting models are increasingly deployed in continuously evolving environments. After a forecast is issued, future observations gradually arrive and reveal how the model has failed. Existing methods mainly treat such observations as supervision for parameter updates. However, it remains unclear whether verified forecasting errors can themselves provide causal feedback for correcting future predictions. To address this, we introduce Error Feedback Calibration (EFC), a fully test-time framework for deterministic post-hoc correction of time series forecasting. EFC constructs a Causal Error Diagonal (CED) from revealed errors of overlapping historical forecasts, providing a fresh view of the frozen forecaster's recent failure behavior. Then Cumulative Error Transport (CET) maps this error state to future residuals using cumulative estimation, while temporal out-of-sample validation controls the correction strength. Temporal Residual Learning (TRL) further predicts only the residual structure left after CET from recent CED trajectories, rather than relearning the entire forecasting error. Extensive experiments across forecasting backbones and datasets show that EFC broadly improves forecasting performance while substantially controlling harmful over correction.

## Residual-Tail-Informed Flow Matching for Time Series Anomaly Detection

DevFeed: [Residual-Tail-Informed Flow Matching for Time Series Anomaly Detection](<https://devfeed.tech/articles/residual-tail-informed-flow-matching-for-time-series-anomaly-detection-137563.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20334>)

Author: Shyam Sundar Saravanan, Tie Luo, Sanjay Madria

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [AI Models](<https://devfeed.tech/topics/ai-models.md>), [VAE](<https://devfeed.tech/topics/vae.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [anomaly-detection](<https://devfeed.tech/tags/anomaly-detection.md>), [autoencoder](<https://devfeed.tech/tags/autoencoder.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [flow-matching](<https://devfeed.tech/tags/flow-matching.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

RT-FM targets false alarms on rare but valid time-series regimes that data-space flow matching may score like true anomalies. It uses a temporal variational autoencoder to estimate residual latent tailness, then turns that estimate into bounded, warmup-annealed weights for training an unsupervised data-space flow-matching model. Across seven TSAD benchmarks, the authors report better rare-normal coverage, fewer false alarms on those windows, and improved aggregate detection while preserving strong anomaly discrimination.

### Source excerpt

arXiv:2611.20334v1 Announce Type: cross Abstract: Time series anomaly detection (TSAD) models are often trained on nominal data using average-risk objectives, which emphasize dominant normal patterns. However, real time series can contain rare but valid nominal regimes that are underrepresented in the training distribution. We show that data-space flow matching can over-penalize such rare-normal windows and assign them anomaly scores comparable to those of true anomalies. To address this, we propose Residual-Tail-Informed Flow Matching (RT-FM), an unsupervised framework that uses a temporal variational autoencoder to estimate residual latent tailness and converts it into bounded, warmup-annealed weights for training a data-space flow matching model. Experiments on seven TSAD benchmarks show that RT-FM improves rare-normal coverage while preserving strong anomaly discrimination, substantially reducing false alarms on such windows and resulting in improved aggregate detection performance. The code is available at: https://github.com/trustAIoT/RT-FM_time-series-anomaly

## Breaking feature collapse in self-supervised time series encoders

DevFeed: [Breaking feature collapse in self-supervised time series encoders](<https://devfeed.tech/articles/breaking-feature-collapse-in-self-supervised-time-series-encoders-137639.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20981>)

Author: \\"Ozg\\"un Turgut, Daniel Rueckert

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [code](<https://devfeed.tech/tags/code.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data-modelling](<https://devfeed.tech/tags/data-modelling.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [distillation](<https://devfeed.tech/tags/distillation.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [rank](<https://devfeed.tech/tags/rank.md>), [scale](<https://devfeed.tech/tags/scale.md>), [similarity](<https://devfeed.tech/tags/similarity.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

The authors attribute poor scaling in general-purpose time-series encoders to partial collapse of output tokens under masked data modeling: input-space reconstruction alone does not constrain redundant feature dimensions. OTISv2 adds feature-space self-distillation against an exponential-moving-average teacher and a KoLeo regularizer alongside reconstruction. In matched training, it reportedly lowers token cosine similarity, raises effective rank, and scales monotonically from 7.1M to 116M parameters, while its predecessor declines. Across 143 uni- and multivariate datasets, it matches or surpasses existing general-purpose encoders. The excerpt says code and pretrained weights are released.

### Source excerpt

arXiv:2611.20981v1 Announce Type: cross Abstract: General-purpose time series encoders promise a single backbone that generalises across tasks and domains; however, they currently do not scale. We trace this down to a critical point of failure: under the predominant masked data modelling supervision, the encoder's output tokens partially collapse. This is the natural consequence of supervising encoders only in the input space, which leaves no constraint against redundant token dimensions in the feature space. We introduce OTISv2, which suppresses the collapse by complementing the reconstruction objective in input space with two objectives in the feature space: a self-distillation loss against an exponential-moving-average teacher, and a KoLeo regulariser. Under matched training, the resulting encoder reduces the cosine similarity and lifts the effective rank of its output tokens relative to its predecessor, and it scales monotonically from 7.1 M to 116 M parameters where its predecessor declines. It matches or surpasses existing general-purpose time series encoders across 143 uni- and multi-variate datasets. To foster further research, we release our code and pre-trained weights.

## Ensemble Modeling for Time Series Forecasting: an Adaptive Robust Optimization Approach

DevFeed: [Ensemble Modeling for Time Series Forecasting: an Adaptive Robust Optimization Approach](<https://devfeed.tech/articles/ensemble-modeling-for-time-series-forecasting-an-adaptive-robust-optimization-approach-138704.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2304.04308>)

Author: L\\'eonard Boussioux, Henry Mao, Dimitris Bertsimas

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [linear-regression](<https://devfeed.tech/topics/linear-regression.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Operations research and optimization](<https://devfeed.tech/topics/operations-research-and-optimization.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [air-pollution](<https://devfeed.tech/tags/air-pollution.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [energy-consumption](<https://devfeed.tech/tags/energy-consumption.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [linear](<https://devfeed.tech/tags/linear.md>), [linear-regression](<https://devfeed.tech/tags/linear-regression.md>), [math-oc](<https://devfeed.tech/tags/math-oc.md>), [methodology](<https://devfeed.tech/tags/methodology.md>), [model-weights](<https://devfeed.tech/tags/model-weights.md>), [modeling](<https://devfeed.tech/tags/modeling.md>), [models](<https://devfeed.tech/tags/models.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [regression](<https://devfeed.tech/tags/regression.md>), [risk](<https://devfeed.tech/tags/risk.md>), [robust](<https://devfeed.tech/tags/robust.md>), [robustness](<https://devfeed.tech/tags/robustness.md>), [shift](<https://devfeed.tech/tags/shift.md>), [synthetic](<https://devfeed.tech/tags/synthetic.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [time-series-forecasting](<https://devfeed.tech/tags/time-series-forecasting.md>)

### AI overview

The paper uses Adaptive Robust Optimization (ARO) to build linear regression ensembles with weights that adapt over time as time-series data shift. In real-world forecasting applications for air pollution, energy consumption, and tropical cyclone intensity, the method reports 16-26% lower RMSE and 14-28% lower 15% conditional value at risk than the best ensemble member in hindsight, and improves on other compared ensemble methods on most metrics. Across eight synthetic disturbance scenarios, it has lower mean RMSE than rolling ridge in seven, with statistically significant improvements in four. A limitation is that an abrupt regime shift appearing only after model selection remains difficult for the approach.

### Source excerpt

arXiv:2304.04308v2 Announce Type: replace Abstract: Accurate time-series forecasting is critical for many problems involving temporal data. Ensemble modeling is a well-established technique that leverages multiple predictive models to improve accuracy and robustness, since a single predictor's performance can vary widely due to shifts in the underlying data distribution. This paper proposes a new methodology for building robust ensembles of time series forecasting models. Our approach uses Adaptive Robust Optimization (ARO) to construct a linear regression ensemble whose model weights adapt over time. We demonstrate the effectiveness of our method through synthetic experiments covering eight disturbance scenarios, from no drift to abrupt regime shifts, and real-world applications, including air pollution management, energy consumption forecasting, and tropical cyclone intensity forecasting. In the real-world applications, our adaptive ensembles outperform the best ensemble member in hindsight by 16-26% in root mean square error (RMSE) and 14-28% in conditional value at risk at the 15% level and improve over the other ensemble methods we compare on most metrics. In the synthetic experiments, they have lower mean RMSE than a rolling-ridge comparator in seven of the eight scenarios, significantly in four, while an abrupt regime shift that first appears after model selection remains a limitation.

## TRACK: Tool-Augmented Reasoning over Time Series for Fault Diagnosis in Power Management System

DevFeed: [TRACK: Tool-Augmented Reasoning over Time Series for Fault Diagnosis in Power Management System](<https://devfeed.tech/articles/track-tool-augmented-reasoning-over-time-series-for-fault-diagnosis-in-power-management-system-137664.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21178>)

Author: Erkan Bayram, Debanjan Ghosh, Li Zhu, Sambarta Dasgupta

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [cs-sy](<https://devfeed.tech/tags/cs-sy.md>), [diagnosis](<https://devfeed.tech/tags/diagnosis.md>), [eess-sy](<https://devfeed.tech/tags/eess-sy.md>), [language-model](<https://devfeed.tech/tags/language-model.md>), [on-device](<https://devfeed.tech/tags/on-device.md>), [power-management](<https://devfeed.tech/tags/power-management.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [sensor](<https://devfeed.tech/tags/sensor.md>), [signal](<https://devfeed.tech/tags/signal.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [tool](<https://devfeed.tech/tags/tool.md>)

### AI overview

TRACK pairs a time-series encoder with a Small Language Model to turn power-system current and voltage waveforms into structured, physics-grounded fault reports. It uses tools for exact, verifiable calculations and runs on-device, keeping proprietary sensor data local and avoiding costly API calls. Its rubric scores report structure, signal fidelity, physics reasoning, diagnosis, and actionability. On a simulated 15-class multi-load benchmark, TRACK exceeded 95% fault-category accuracy and generalized to out-of-distribution faults; the best encoder-probing baseline achieved 58%. The supplied abstract does not detail the rubric results or the scope of out-of-distribution testing.

### Source excerpt

arXiv:2611.21178v1 Announce Type: cross Abstract: Fault diagnosis in power management systems running multiple loads relies on experts manually inspecting current and voltage waveforms, which fails to scale across deployed units. We present TRACK, which pairs a time series encoder with a Small Language Model to produce structured, physics-grounded diagnostic reports, calling tools for exact, verifiable calculations. It operates on-device, keeping proprietary sensor data local and avoiding expensive API calls. We introduce a rubric-based framework that evaluates these reports by scoring structure, signal fidelity, physics reasoning, diagnosis, and actionability through per-criterion verdicts. On a 15-class multi-load simulated benchmark, TRACK achieves over 95% fault-category accuracy and generalizes to out-of-distribution faults, whereas the best encoder-probing baseline reaches only 58% accuracy.

## Illusion of Explainability: A Sanity Check for Spectral Band Attribution in Satellite Time-Series Crop Classification

DevFeed: [Illusion of Explainability: A Sanity Check for Spectral Band Attribution in Satellite Time-Series Crop Classification](<https://devfeed.tech/articles/illusion-of-explainability-a-sanity-check-for-spectral-band-attribution-in-satellite-time-series-crop-classification-138955.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19889>)

Author: Sumit Tahiliani, Avi Mathur, Sandarbh Agarwal, Rakshith Sathish, Ashwathi Nambiar

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Feature Engineering](<https://devfeed.tech/topics/feature-engineering.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [reproducibility](<https://devfeed.tech/topics/reproducibility.md>)

Tags: [attribution](<https://devfeed.tech/tags/attribution.md>), [band](<https://devfeed.tech/tags/band.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [classification](<https://devfeed.tech/tags/classification.md>), [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [crop](<https://devfeed.tech/tags/crop.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [explainability](<https://devfeed.tech/tags/explainability.md>), [feature](<https://devfeed.tech/tags/feature.md>), [model](<https://devfeed.tech/tags/model.md>), [performance](<https://devfeed.tech/tags/performance.md>), [rankings](<https://devfeed.tech/tags/rankings.md>), [redundancy](<https://devfeed.tech/tags/redundancy.md>), [remote-sensing](<https://devfeed.tech/tags/remote-sensing.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [satellite](<https://devfeed.tech/tags/satellite.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

On the PASTIS Sentinel-2 crop-classification task, plausible spectral-band attribution rankings did not reliably identify bands whose use improved retrained performance. GradientSHAP, Shapley sampling, and Occlusion highlighted vegetation-sensitive regions, but deletion and insertion results for Shapley and Occlusion were sensitive to feature ordering. Aggregate agreement also concealed per-class differences: pairwise mean Spearman correlations ranged from 0.09 to 0.79. Retraining from scratch with the five lowest-ranked bands performed comparably to or better than using the five highest-ranked bands, and a separately trained model with the same configuration and seed produced a different top-five subset with better retraining performance. Across all 252 five-band subsets, effective dimensionality was low (1.35-2.62), consistent with spectral redundancy. The abstract cautions that plausible attribution rankings in this setting do not establish which individual bands are task-relevant.

### Source excerpt

arXiv:2611.19889v1 Announce Type: cross Abstract: Attribution methods are commonly validated through convergence across methods and consistency with domain knowledge, yet neither criterion establishes whether identified features are necessary for a task. This limitation is particularly relevant in multispectral remote sensing, where overlapping spectral information makes individual bands potentially substitutable. We investigate this gap in crop-type classification using GradientSHAP, Shapley sampling, and Occlusion on a U-TAE segmentation model trained on Sentinel-2 time series from the PASTIS benchmark. All three methods identify vegetation-sensitive spectral regions consistent with domain knowledge, while Shapley and Occlusion exhibit strong ordering sensitivity in deletion and insertion tests. However, these properties do not reliably predict retraining performance. Aggregate attribution consensus masks substantial per-class disagreement, with pairwise mean Spearman correlations ranging from 0.09 to 0.79. Models retrained from scratch using the five lowest-ranked bands perform comparably to or better than those using the five highest-ranked bands. Furthermore, attribution rankings from an independently trained model with identical configuration and seed identify a different top-five subset that yields better retraining performance. Analysis of all 252 five-band subsets reveals consistently low effective dimensionality (1.35 to 2.62), suggesting substantial spectral redundancy. These findings demonstrate that plausible and apparently faithful attribution rankings do not necessarily identify task-relevant bands for improved retrained performance, highlighting the limitations of band-level feature attribution in spectrally redundant input spaces.

## FH-Sampling: Frequency-aware Hypergraph-driven Oversampling for Imbalanced Time-Series Classification

DevFeed: [FH-Sampling: Frequency-aware Hypergraph-driven Oversampling for Imbalanced Time-Series Classification](<https://devfeed.tech/articles/fh-sampling-frequency-aware-hypergraph-driven-oversampling-for-imbalanced-time-series-classification-139119.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20991>)

Author: Yunrong Hao, Ruohao Dai, Zhen Liu, Guan Wang, Wenbin Pei, Qiang Zhang

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Time Series](<https://devfeed.tech/topics/time-series.md>), [fourier transform](<https://devfeed.tech/topics/fourier-transform.md>)

Tags: [aircraft](<https://devfeed.tech/tags/aircraft.md>), [applications](<https://devfeed.tech/tags/applications.md>), [classification](<https://devfeed.tech/tags/classification.md>), [components](<https://devfeed.tech/tags/components.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [diagnosis](<https://devfeed.tech/tags/diagnosis.md>), [distribution](<https://devfeed.tech/tags/distribution.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [imbalanced-data](<https://devfeed.tech/tags/imbalanced-data.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [series](<https://devfeed.tech/tags/series.md>), [techniques](<https://devfeed.tech/tags/techniques.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [time-series-data](<https://devfeed.tech/tags/time-series-data.md>)

### AI overview

FH-Sampling addresses imbalanced time-series classification by combining hypergraph-guided sample relationships with frequency-aware synthesis. It builds a hypergraph in the original time-series space to capture higher-order relationships and adaptively assigns sampling ratios to reduce the influence of noisy samples. A discrete wavelet transform separates frequency components; guided by the hypergraph, the method combines high-frequency components from related samples while retaining the target sample's low-frequency component. The abstract reports advantages on 24 imbalanced datasets and says tests on simulated aircraft trajectories produced trajectories that better matched the real distribution. It does not provide metrics or details of the comparisons.

### Source excerpt

arXiv:2611.20991v1 Announce Type: new Abstract: Imbalanced time-series data appears commonly in diverse applications, such as medical diagnosis. When trained on imbalanced data, classifiers tend to exhibit a bias toward the majority class, making it difficult to identify minority-class instances accurately, despite their critical importance. Sampling techniques have proven effective in rebalancing imbalanced data. However, existing interpolation-based oversampling methods typically rely on pairwise interpolation, limiting their ability to capture high-order inter-sample relationships, and their inherent smoothing effects tend to suppress high-frequency components, compromising spectral fidelity. Therefore, we propose a Frequency-aware Hypergraph-driven Oversampling method (FH-Sampling) for imbalanced time-series classification. In FH-Sampling, a hypergraph is first constructed in the original time-series space to capture complex high-order inter-sample relationships, and sampling ratios are adaptively assigned to mitigate the impact of noisy samples. The discrete wavelet transform decomposes each sample into frequency components. Guided by the sample hypergraph, FH-Sampling combines high-frequency components from related samples while retaining the target sample's low-frequency component to help preserve important frequency characteristics. Experiments on 24 imbalanced datasets demonstrate the advantages of FH-Sampling for imbalanced time-series classification. We further apply FH-Sampling to aircraft trajectory generation. Experiments on simulated aircraft trajectories show that FH-Sampling produces trajectories that better match the real distribution, highlighting its potential for practical applications.

## Time series causal discovery with variable lags

DevFeed: [Time series causal discovery with variable lags](<https://devfeed.tech/articles/time-series-causal-discovery-with-variable-lags-134977.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2605.04081>)

Author: Bruno Petrungaro, Anthony C. Constantinou

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [bayesian](<https://devfeed.tech/tags/bayesian.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [covid-19](<https://devfeed.tech/tags/covid-19.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [discovery](<https://devfeed.tech/tags/discovery.md>), [lag](<https://devfeed.tech/tags/lag.md>), [learning](<https://devfeed.tech/tags/learning.md>), [parallel](<https://devfeed.tech/tags/parallel.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [score](<https://devfeed.tech/tags/score.md>), [search](<https://devfeed.tech/tags/search.md>), [simulations](<https://devfeed.tech/tags/simulations.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [uk](<https://devfeed.tech/tags/uk.md>), [variable](<https://devfeed.tech/tags/variable.md>)

### AI overview

The method jointly learns a temporal ordering and one dominant lag for each selected ordered variable pair, using a decomposable BIC-style score for mixed data that accounts for node-specific sample sizes and favours recent lags. Tabu search with change-lag moves and greedy per-child lag tuning optimises the score; node-wise decomposition enables cached and parallel neighbourhood evaluation. Simulations found it outperformed a fixed lag-1 score-based learner in most evaluated settings, while most specialised methods did better when their assumptions held. On UK COVID-19 data, 46.8% of selected edges had lags above one despite an overall short-lag skew. The approach assumes no contemporaneous edges and causal sufficiency; recovery can be limited when effects span multiple lags.

### Source excerpt

arXiv:2605.04081v3 Announce Type: replace-cross Abstract: Causal Bayesian Networks (CBNs) are a powerful tool for reasoning under uncertainty about complex real-world problems. Such problems evolve over time, responding to external shocks as they occur. To support decision-making, CBNs require a cause-and-effect map of the variables under consideration, known as the network's structure. Learning the graphical structure of a causal model from data remains challenging; learning it from time-series data is even harder because dependencies may arise at different time lags. We construct a method that jointly learns a temporal ordering and a dominant lag for each selected ordered variable pair. Our contribution is a decomposable BIC-style objective score for mixed data that combines node-specific sample sizes with a penalty favouring recent lags. We optimise this objective using a Tabu search augmented with change-lag moves and greedy per-child lag tuning, while its node-wise decomposition supports cached and parallel neighbourhood evaluation. We establish acyclicity and complexity results and derive recovery limits when effects are distributed across multiple lags. In simulations, we show that the proposed method outperforms an inflexible lag-1 score-based learner in most evaluated settings, whereas most specialised methods achieve higher recovery accuracy on data that satisfy their assumptions. On a real-world UK COVID-19 dataset, the method learnt an expected short-lag-skewed structure; nevertheless, 46.8% of selected edges had lags greater than one, showing that longer delays can be retained when their improved fit offsets the penalty. Under its no-contemporaneous-edges and causal-sufficiency assumptions, the method provides a computationally tractable subroutine for edge-specific single-lag optimisation, complementing score-based causal discovery methods.

## T2ST: Data-Efficient Spatio-Temporal Foundation Models via Temporal Data and Knowledge Transfer

DevFeed: [T2ST: Data-Efficient Spatio-Temporal Foundation Models via Temporal Data and Knowledge Transfer](<https://devfeed.tech/articles/t2st-data-efficient-spatio-temporal-foundation-models-via-temporal-data-and-knowledge-transfer-135175.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18199>)

Author: Yue Xu, Kaiyun Wei, Wenying Duan, Zichen Xu, Shuai Ma, Xiaoxi He

Published: 2026-11-16T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Predictive AI](<https://devfeed.tech/topics/predictive-ai.md>), [Post-training optimization](<https://devfeed.tech/topics/post-training-optimization.md>)

Tags: [architectures](<https://devfeed.tech/tags/architectures.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [dependency](<https://devfeed.tech/tags/dependency.md>), [distillation](<https://devfeed.tech/tags/distillation.md>), [external](<https://devfeed.tech/tags/external.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [framework](<https://devfeed.tech/tags/framework.md>), [knowledge-transfer](<https://devfeed.tech/tags/knowledge-transfer.md>), [models](<https://devfeed.tech/tags/models.md>), [patches](<https://devfeed.tech/tags/patches.md>), [pattern](<https://devfeed.tech/tags/pattern.md>), [performance](<https://devfeed.tech/tags/performance.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [resolutions](<https://devfeed.tech/tags/resolutions.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [series](<https://devfeed.tech/tags/series.md>), [spatial](<https://devfeed.tech/tags/spatial.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [temporal-data](<https://devfeed.tech/tags/temporal-data.md>)

### AI overview

T2ST aims to reduce the spatio-temporal data needed to train spatio-temporal foundation models (STFMs). It augments scarce observations with selected auxiliary time series and transfers knowledge from a frozen time-series foundation model: matched sequences reinforce sparsely observed temporal patterns, complementary sequences broaden temporal coverage, and temporal-relation and multi-scale prediction distillation transfer temporal structure and forecasting behavior across heterogeneous architectures. Spatial dependency learning remains grounded in real spatio-temporal observations. The abstract reports that training with 10% of the original spatio-temporal data achieved overall forecasting performance competitive with or better than full-data counterparts across six long-term and twelve short-term datasets.

### Source excerpt

arXiv:2611.18199v1 Announce Type: cross Abstract: Spatio-temporal foundation models (STFMs) require large-scale training corpora, yet synchronized observations and spatial metadata are costly to acquire and curate, limiting the availability of suitable data. By contrast, time-series data are abundant and easier to collect. Since observations at each location form a time series, this motivates a natural question: can abundant time-series data enhance STFMs when spatio-temporal observations are scarce? To bridge this gap, we propose T2ST, a data-efficient training framework that augments limited spatio-temporal observations with selected auxiliary time series and transfers knowledge from a frozen time-series foundation model. Our key insight is that temporal pattern learning can benefit from abundant external time series, while spatial dependency learning remains grounded in real spatio-temporal observations. Guided by this insight, T2ST retrieves matched sequences to reinforce sparsely observed temporal patterns and selects reliable complementary sequences to broaden temporal coverage. Temporal relation distillation aligns relationships among historical patches, while multi-scale prediction distillation transfers forecasting behavior at fine and coarse temporal resolutions, enabling knowledge transfer across heterogeneous architectures. Using only 10\% of the original spatio-temporal data, T2ST enables STFMs to achieve overall forecasting performance competitive with or superior to their full-data counterparts across six long-term and twelve short-term forecasting datasets. These results establish a new paradigm for data-efficient training of spatio-temporal foundation models. Code is available at https://anonymous.4open.science/r/T2ST-65F3/.

## TSseek: Pattern-Based Search for Distributed Time Series Datasets

DevFeed: [TSseek: Pattern-Based Search for Distributed Time Series Datasets](<https://devfeed.tech/articles/tsseek-pattern-based-search-for-distributed-time-series-datasets-132672.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2606.09824>)

Author: Xiaoshuai Li, Khalid Alnuaim, Mohamed Y. Eltabakh, Elke A. Rundensteiner

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [big-data](<https://devfeed.tech/topics/big-data.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [search-engines](<https://devfeed.tech/topics/search-engines.md>)

Tags: [cs-db](<https://devfeed.tech/tags/cs-db.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [engine](<https://devfeed.tech/tags/engine.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [index](<https://devfeed.tech/tags/index.md>), [language](<https://devfeed.tech/tags/language.md>), [matching](<https://devfeed.tech/tags/matching.md>), [metadata](<https://devfeed.tech/tags/metadata.md>), [order](<https://devfeed.tech/tags/order.md>), [pattern](<https://devfeed.tech/tags/pattern.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [plan](<https://devfeed.tech/tags/plan.md>), [processing](<https://devfeed.tech/tags/processing.md>), [queries](<https://devfeed.tech/tags/queries.md>), [query](<https://devfeed.tech/tags/query.md>), [range](<https://devfeed.tech/tags/range.md>), [reading](<https://devfeed.tech/tags/reading.md>), [recall](<https://devfeed.tech/tags/recall.md>), [search](<https://devfeed.tech/tags/search.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

TSseek searches distributed time series for patterns involving trends, value ranges, wildcards, and constraints on changes, duration, position, order, and spacing. It builds TSseek-X from error-bounded line-segment approximations and their slope and range metadata. Query planning uses selectivity estimates; spatial probes find candidates while retaining every true match, then exact refinement checks candidates against original readings. Across four datasets of up to 200 million series (25.6 billion readings), TSseek reports a 20x speedup over a distributed exact full scan using the same matcher. In whole-matching comparisons, adapted baselines reached at most 57% recall, while TSseek returned the exact answer set.

### Source excerpt

arXiv:2606.09824v5 Announce Type: replace Abstract: In applications such as household energy monitoring and ECG analysis, analysts may search for time series that satisfy constraints on trends, value ranges, and durations. To support such queries at scale, we propose TSseek, a distributed processing engine for time series pattern queries. Its regex-inspired language enables users to compose trends, value ranges, and wildcards with constraints on changes between readings and the duration, position, order, and spacing of patterns. During indexing, TSseek approximates each stored time series with error-bounded line segments, records their slope and value-range metadata, and builds TSseek-X, a distributed spatial index over the segments. At query time, TSseek rewrites the query and uses selectivity estimates to plan the index lookup. During execution, index lookup uses spatial probes derived from query constraints over reading positions and values to select candidate series from TSseek-X while guaranteeing to retain every true match. Finally, exact refinement checks each candidate against the full query using its original readings. Experiments on four datasets of up to 200 million time series (25.6 billion readings) show 20x speedup over a distributed exact full scan using the same matcher. In the whole-matching comparison, the adapted baselines each reach at most 57% recall, while TSseek returns the exact answer set.

## CoLaX: Context-Aware Local Explanations for Time Series Classification

DevFeed: [CoLaX: Context-Aware Local Explanations for Time Series Classification](<https://devfeed.tech/articles/colax-context-aware-local-explanations-for-time-series-classification-132090.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15067>)

Author: Kexuan Zhang, Paolo Frazzetto, Xiaobei Zou, Gustau Camps-Valls, Yang Tang

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Neural Network](<https://devfeed.tech/topics/neural-network.md>)

Tags: [attribution](<https://devfeed.tech/tags/attribution.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [classification](<https://devfeed.tech/tags/classification.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [ecg](<https://devfeed.tech/tags/ecg.md>), [global-context](<https://devfeed.tech/tags/global-context.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [sequence](<https://devfeed.tech/tags/sequence.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

CoLaX addresses a limitation of local-pattern explanations: the same time-series motif can matter differently depending on sequence-wide trends or regimes. It combines shapelet-driven Shapley attribution with a learned full-sequence representation that supplies sample-specific prototype weights and a temporal signal to reweight local activations. The resulting context-conditioned segments are evaluated as Shapley players using only the frozen black-box classifier. On synthetic benchmarks with context-dependent motifs, CoLaX improves AUPRC by an average of 30% over the strongest baseline; on real-world data, it reports the highest AUPRC on ECG. Ablations attribute gains to input-dependent conditioning rather than extra learnable weights. The abstract suggests it is most useful when recurring motifs change relevance across sequence contexts.

### Source excerpt

arXiv:2611.15067v1 Announce Type: cross Abstract: Post-hoc explanations for time series classifiers usually score local patterns, such as subsequences or shapelets, without regard to the rest of the sequence. Yet the same motif can carry different predictive meaning under different global temporal conditions, such as trends or regimes. We propose CoLaX, a context-aware post-hoc explanation framework that builds on shapelet-driven Shapley attribution and conditions local-pattern relevance on a learned representation of the full sequence. Global context acts at two levels: it produces sample-specific prototype weights and a temporally resolved signal that reweights the local activation map. The resulting context-conditioned segments are the players of a Shapley attribution game whose values come only from the frozen black-box classifier. On synthetic benchmarks where identical motifs change relevance across contexts, CoLaX improves AUPRC over the strongest baseline by 30\% on average. On real-world data, it achieves the highest AUPRC on ECG. Additional ablations indicate that the gains come from input-dependent conditioning rather than from additional learnable weights. CoLaX is therefore most useful when recurring local motifs change relevance across sequence-level contexts.

## PRELUDE: Separating Common Components from Target-Specific Dependencies for Multivariate Time Series Forecasting

DevFeed: [PRELUDE: Separating Common Components from Target-Specific Dependencies for Multivariate Time Series Forecasting](<https://devfeed.tech/articles/prelude-separating-common-components-from-target-specific-dependencies-for-multivariate-time-series-forecasting-132964.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.15155>)

Author: Dongqing Song, Jianyu Xu, Haobin Ding, Xinyi Guo, Jiahao Du, Ruiqi He, Ruiqi Liu, Yuntong Xie, Chao He

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [attention](<https://devfeed.tech/tags/attention.md>), [components](<https://devfeed.tech/tags/components.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [framework](<https://devfeed.tech/tags/framework.md>), [graph](<https://devfeed.tech/tags/graph.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

The authors find that cross-variable forecasting modules can reduce error while reading nearly the same inputs for every target, with much of the gain attributed to shared common components rather than target-specific dependencies. They call this spurious dependency, linking it to common-component confounding that can obscure dependencies and incompletely capture common factors. PRELUDE separates these tasks: it estimates multiple common factors through cross-sectional pooling, then reads target-specific lead-lag dependencies in a prewhitened innovation space. In experiments on 16 datasets across four domains, it reportedly achieved the best MSE in 23 of 24 short-horizon and air-quality settings and lowered average MSE by 6.1% versus the strongest baseline.

### Source excerpt

arXiv:2611.15155v1 Announce Type: new Abstract: Multivariate time series forecasting often improves the prediction of a target by modeling the interactions among variables, and cross-variable modules such as cross-sectional pooling, inter-variable attention, and graph structures are used to construct such interactions. However, by intervening on the cross-variable reading of existing models at inference time, we find that although these models substantially reduce forecasting error through cross-variable interaction, their reading is nearly identical for all targets: the gain appears to come from dependencies among variables but mainly comes from the common component. We term this phenomenon spurious dependency and attribute it to common-component confounding: present in every variable, the common component affects how reading distinguishes sources, so that reading tends to be shared across targets and concentrates on a few similar averages; dependencies thus make up a small share of the read-in content, and a common component composed of several drivers is only partly captured. Theoretical analysis further shows that the common component and dependencies place different requirements on reading and should therefore be handled in different ways. Accordingly, we propose PRELUDE, a separated-reading framework that estimates the common component holistically through cross-sectional pooling and multiple common factors, and reads dependencies for each target via lead-lag relations in the innovation space obtained by prewhitening. Extensive experiments on 16 real-world datasets from four domains show that PRELUDE effectively mitigates spurious dependency and achieves the best overall performance: it attains the best MSE on 23 of the 24 short-horizon and air-quality settings and reduces the average MSE by 6.1% relative to the strongest baseline.

## Learning Temporal Dynamics with Derivative Representations for Time Series Forecasting

DevFeed: [Learning Temporal Dynamics with Derivative Representations for Time Series Forecasting](<https://devfeed.tech/articles/learning-temporal-dynamics-with-derivative-representations-for-time-series-forecasting-133123.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16514>)

Author: Shujie Duan, Jia Hong Puah, Haocheng Wu, Yifei Mao, Hongting Zhou, Chenxuan Li, Zitao Fang, Yang Lu, Sim Kuan Goh

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Physics-guided deep learning](<https://devfeed.tech/topics/physics-guided-deep-learning.md>)

Tags: [coherence](<https://devfeed.tech/tags/coherence.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [deep-learning](<https://devfeed.tech/tags/deep-learning.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [learning](<https://devfeed.tech/tags/learning.md>), [model](<https://devfeed.tech/tags/model.md>), [model-agnostic](<https://devfeed.tech/tags/model-agnostic.md>), [series](<https://devfeed.tech/tags/series.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [time-series-forecasting](<https://devfeed.tech/tags/time-series-forecasting.md>), [training](<https://devfeed.tech/tags/training.md>), [transformers](<https://devfeed.tech/tags/transformers.md>)

### AI overview

DeltaTSF trains forecasting models to jointly represent time series and their discrete temporal derivatives across input and forecast horizons. Coupling these representations is intended to encourage temporal consistency, described as a form of discrete Taylor consistency, and produce more stable, dynamically coherent forecasts. The framework is model-agnostic, requires no explicit derivative computation, and adds no inference-time computational overhead. The abstract reports consistent performance improvements and more coherent predictions across diverse benchmarks and backbone architectures, but provides no specific results or benchmark details. Source code is stated to be forthcoming upon acceptance.

### Source excerpt

arXiv:2611.16514v1 Announce Type: new Abstract: Time series forecasting aims to predict the future evolution of complex systems by learning their underlying dynamics from past observations. Despite recent advances in deep learning, Transformers in particular still struggle to capture the temporal evolution of system dynamics. Existing approaches have been largely driven by architectural innovation, with comparatively little attention paid to training strategies that explicitly exploit the temporal structure governing system dynamics. In this work, we introduce DeltaTSF, a dynamics-aware forecasting framework based on discrete derivative representations of time series. Specifically, DeltaTSF jointly learns the series and their discrete temporal derivatives across both input and forecast horizons within a unified, model-agnostic framework. By coupling these representations over time, the model is encouraged to maintain temporal consistency, a property we show to induce a form of discrete Taylor consistency during training. This formulation enables the model to better reflect the underlying system dynamics, resulting in more stable and physically coherent forecasts. Notably, DeltaTSF does not require explicit derivative computation and introduces no additional computational overhead at inference time. Extensive experiments across diverse benchmarks and backbone architectures demonstrate that DeltaTSF consistently improves forecasting performance and yields more dynamically coherent predictions than strong SOTA methods. These results highlight the effectiveness of enforcing dynamic coherence through derivative representations as a model-agnostic paradigm that operates alongside and complements architectural design. Source code will be made publicly available upon acceptance.

## Closing the Loop: Co-Evolving EM for Irregular Time Series Generation in Lifted Representations

DevFeed: [Closing the Loop: Co-Evolving EM for Irregular Time Series Generation in Lifted Representations](<https://devfeed.tech/articles/closing-the-loop-co-evolving-em-for-irregular-time-series-generation-in-lifted-representations-132900.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.14659>)

Author: Idan Arbiv, Gal Fadlon, Omri Azencot

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [iteration](<https://devfeed.tech/topics/iteration.md>)

Tags: [2d](<https://devfeed.tech/tags/2d.md>), [approach](<https://devfeed.tech/tags/approach.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [context](<https://devfeed.tech/tags/context.md>), [correlation](<https://devfeed.tech/tags/correlation.md>), [corruption](<https://devfeed.tech/tags/corruption.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [diffusion](<https://devfeed.tech/tags/diffusion.md>), [em](<https://devfeed.tech/tags/em.md>), [error](<https://devfeed.tech/tags/error.md>), [feature](<https://devfeed.tech/tags/feature.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generation](<https://devfeed.tech/tags/generation.md>), [irregular](<https://devfeed.tech/tags/irregular.md>), [series](<https://devfeed.tech/tags/series.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

For generating regular time series from irregular observations, the paper argues that one-shot imputation is limited because point estimates average over ambiguous gaps and stochastic imputations can become stale as the generator changes. It proposes co-evolving Monte Carlo EM: each E-step samples missing values from the current generator's posterior, and each M-step retrains the generator on those completions. For a diffusion prior in a lifted 2D representation conditioned on time-series observations, PSLR is designed to maintain operator-consistent conditioning, uncertainty, and manifold consistency. Across ten datasets and four corruption regimes, the authors report gains over the strongest baselines in discriminative, predictive, Context-FID, and feature-correlation metrics; ablations report a 73% reduction in average discriminative score versus single-step baselines. These are abstract-level claims; the supplied excerpt gives no experimental details.

### Source excerpt

arXiv:2611.14659v1 Announce Type: new Abstract: Regular time-series generation from irregular time series is typically reduced to one-shot imputation: complete each partial sequence once, freeze the completions, and train a generator on the resulting surrogate dataset. We show that this open-loop strategy is structurally flawed. Point imputers collapse ambiguous missing regions to conditional averages, while stochastic imputers become stale as the generator evolves; the binding constraint is therefore iteration, not imputer design. We introduce a co-evolving Monte Carlo EM framework that closes the imputer-generator loop. Each E-step samples missing values from the posterior of the current generator, and each M-step retrains the generator on these completions, preserving unconditional generator training. Since our diffusion prior operates in a lifted 2D representation while observations live in time-series space, we further introduce Posterior Sampling for Lifted Representations (PSLR), which enforces operator-consistent conditioning, uncertainty preservation, and manifold consistency during the E-step. Across ten datasets and four corruption regimes, our method achieves state-of-the-art performance, improving over the strongest baselines by 59% in discriminative score, 15% in predictive score, 73% in Context-FID, and 71% in feature-correlation error. Ablations show that our approach reduces average discriminative score by 73% over single-step baselines, closing most of the remaining gap to the clean-data oracle. Code is available at https://github.com/azencot-group/PSLR.

## Time and Space: A Dual-Axis, Evidence-Grounded Context Substrate for Personal Agents

DevFeed: [Time and Space: A Dual-Axis, Evidence-Grounded Context Substrate for Personal Agents](<https://devfeed.tech/articles/time-and-space-a-dual-axis-evidence-grounded-context-substrate-for-personal-agents-132245.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16492>)

Author: Haitao Cheng, Jerry Hu, Longdong Mao, Dong Huang, Dongming Li, Lijuan Zhang, Mengting Zhang

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [context](<https://devfeed.tech/topics/context.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [hybrid-search](<https://devfeed.tech/topics/hybrid-search.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [agents](<https://devfeed.tech/tags/agents.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [axis](<https://devfeed.tech/tags/axis.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [constraint](<https://devfeed.tech/tags/constraint.md>), [context](<https://devfeed.tech/tags/context.md>), [cost](<https://devfeed.tech/tags/cost.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [define](<https://devfeed.tech/tags/define.md>), [device](<https://devfeed.tech/tags/device.md>), [embedded](<https://devfeed.tech/tags/embedded.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [event](<https://devfeed.tech/tags/event.md>), [personal](<https://devfeed.tech/tags/personal.md>), [personal-agents](<https://devfeed.tech/tags/personal-agents.md>), [space](<https://devfeed.tech/tags/space.md>), [substrate](<https://devfeed.tech/tags/substrate.md>), [time](<https://devfeed.tech/tags/time.md>)

### AI overview

The paper proposes spatio-temporal context (STC) as a way for personal agents to track which entities facts concern, when they apply, and which source records support them. Its temporal axis represents event time, validity intervals, and ingestion time; its relational axis connects entities through typed, semantically measured links. Relations are anchored to immutable records, and answers must remain traceable to those records. The framework includes ambiguity-preserving entity resolution, conflict detection that sends mutually exclusive facts to user adjudication, and four use regimes: query, insight, triage, and proactive reminders. A local-first retrieval prototype uses SQLite, on-device embeddings, and hybrid lexical-dense recall; reported store measurements are 341 documents, 340,079 chunks, and 190,271 embedded vectors. The analytic layer is specified but not implemented, and the paper proposes evaluation metrics including evidence coverage, temporal correctness, conflict precision, and interruption cost.

### Source excerpt

arXiv:2611.16492v1 Announce Type: new Abstract: We argue that a binding constraint for personal agents is not model capacity but the form in which context is represented: the agent does not know which entity a phrase denotes, when a fact held, or which source record licenses a claim. We call the missing representation spatio-temporal context (STC) and define it on two axes. The temporal axis carries event time, validity intervals and ingestion time. The spatial axis is a relational space in which entities are points, typed relations are edges, and distance is semantic rather than geographic; physical places are one entity type among many. Every relation is anchored to immutable source records, and every answer must be closed over those records. We formalise the substrate and six invariants, a context operator that composes temporal intersection with relational expansion, an ambiguity-preserving resolution state machine, a conflict detector that converts mutually exclusive facts into user adjudication rather than silent overwrite, and a four-regime consumption taxonomy spanning query, insight, triage and proactive reminder. We describe a local-first prototype of the retrieval layer built on SQLite with on-device embeddings and hybrid lexical-dense recall, report measurements from a deployed store of 341 documents, 340,079 chunks and 190,271 embedded vectors, and specify, but do not yet implement, the analytic layer on which the framework depends. We close with an evaluation protocol with per-invariant metrics for evidence coverage, temporal correctness, conflict precision and interruption cost. The proposal targets the representation of context rather than the volume of it.

## Quantifying the Sampling Frequency Invariance of Time-Series Foundation Models for Sensor Data Applications

DevFeed: [Quantifying the Sampling Frequency Invariance of Time-Series Foundation Models for Sensor Data Applications](<https://devfeed.tech/articles/quantifying-the-sampling-frequency-invariance-of-time-series-foundation-models-for-sensor-data-applications-128432.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.14425>)

Author: Sourish Gunesh Dhekane, Abinaya Kumar, Torsten Woertwein, Yan-Bo Lin, Thomas Ploetz, Shabnam Ghaffarzadegan

Published: 2026-11-11T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [invariance](<https://devfeed.tech/tags/invariance.md>), [performance](<https://devfeed.tech/tags/performance.md>), [property](<https://devfeed.tech/tags/property.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [sensor](<https://devfeed.tech/tags/sensor.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [time-series-foundation-models](<https://devfeed.tech/tags/time-series-foundation-models.md>), [used](<https://devfeed.tech/tags/used.md>)

### AI overview

The study evaluates how robust four time-series foundation models--MOMENT, Chronos-2, Moirai-MoE, and TimesFM--are to changes in sensor sampling frequency. It measures embedding shifts and declines in forecasting and downstream performance using three proposed metrics: FSI, FDI, and CDI. The abstract says invariance depends on how much bandwidth remains for the task relative to the signal's native sampling frequency, and examines how model choice and information loss contribute to performance changes. It frames sampling-frequency invariance as a measurable property of a signal-model pair; the supplied excerpt does not report model-specific results.

### Source excerpt

arXiv:2611.14425v1 Announce Type: cross Abstract: Time-Series Foundation Models (TSFMs) are increasingly used in sensor applications, where sampling frequency (SF) varies across sensors, devices, and operating modes (e.g., power saving). Ideally, TSFMs must remain effective during such SF changes, provided the SF retains task-critical bandwidth. In this paper, we stress-test 4 prominent TSFMs: MOMENT, Chronos-2, Moirai-MoE, and TimesFM, to quantify their SF-invariance and investigate whether their behavior tracks the resulting loss of information bandwidth rather than the SF value itself. Our stress-test probes 3 dimensions: (i) the shift in embeddings, (ii) the degradation in forecasting performance, and (iii) the degradation in downstream performance, quantified by three novel metrics: FSI, FDI, and CDI, respectively. To our knowledge, this is the first study to place SF-invariance at the core of TSFM evaluations. Our results characterize: (i) how SF-invariance depends on the margin between signal's native-SF and the bandwidth its task requires, (ii) how much of that invariance the choice of TSFM explains, and (iii) how much of TSFM's performance degradation under a lower SF reflects information loss rather than a limitation of their abilities. These findings reframe SF-invariance as a measurable, signal-aware property of the (signal X TSFM) pair, and guide practitioners on when TSFMs display SF-invariance.

## INTENT: Out-of-Sample, Out-of-Time, and Cold-Start Evaluation on User Event Sequences

DevFeed: [INTENT: Out-of-Sample, Out-of-Time, and Cold-Start Evaluation on User Event Sequences](<https://devfeed.tech/articles/intent-out-of-sample-out-of-time-and-cold-start-evaluation-on-user-event-sequences-128652.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.13069>)

Author: Mattias Kvist, Erik Lidman Hillbom, Edoardo de Cal, Davis Siemens, Adam Fredriksson, Rickard Br\\"uel-Gabrielsson

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Dataset development](<https://devfeed.tech/topics/dataset-development.md>), [Sequences](<https://devfeed.tech/topics/sequences.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [cold-start](<https://devfeed.tech/tags/cold-start.md>), [cs-ir](<https://devfeed.tech/tags/cs-ir.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [event](<https://devfeed.tech/tags/event.md>), [sequences](<https://devfeed.tech/tags/sequences.md>), [time](<https://devfeed.tech/tags/time.md>), [user](<https://devfeed.tech/tags/user.md>)

### AI overview

INTENT benchmarks next-action prediction across seven cells defined by whether the user, time period, and target item appeared in training. An audit of 59 public dataset families retained five; missing calendar timestamps were the most common defect, and each dataset uses a cutoff chosen by a rule set before model evaluation. In reference results, a transformer performed best on seen items after the cutoff but scored zero on new items. Item metadata helped mainly on MicroLens, while repeating the user's last item led on new items in four datasets. That parameter-free baseline indicates that on three datasets most post-cutoff score loss reflects changed behavior rather than model failure. The benchmark pipeline, cutoff rule, and per-cell user counts are public.

### Source excerpt

arXiv:2611.13069v1 Announce Type: new Abstract: Banks now train foundation models on ordered sequences of customer events such as transactions, app sessions, and authorizations. Deployed, such a model predicts forward in time, often for customers and merchants absent from training, yet each is evaluated on private data, so no two can be compared. Public benchmarks for such sequences rarely test a period the model has not seen; recommendation studies usually hold out each user's last events instead. We present INTENT (INTEraction, Novelty, and Transfer evaluation for next-action prediction), a public benchmark that scores a model in seven cells defined by whether the user, the period, and the target item appeared in training. An audit of 59 public dataset families retains five datasets; the most common defect is the lack of a calendar timestamp that orders events. No date lies within the time span of all five datasets, so each is split at its own cutoff, chosen by one rule fixed before any model is evaluated. Summed over the datasets, every cell holds at least 111,769 users. In reference results the cell determines the best model: a transformer leads on seen items after the cutoff but scores zero on new items, a variant that embeds item metadata helps mainly on MicroLens, and repeating the user's last item leads on new items in four of five datasets. This parameter-free control also shows that on three datasets most of the score lost after the cutoff reflects changed behavior rather than the model. The pipeline, the cutoff rule, and the per-cell user counts are publicly available.

## Time Series Change Point Detection with Gaussian Processes

DevFeed: [Time Series Change Point Detection with Gaussian Processes](<https://devfeed.tech/articles/time-series-change-point-detection-with-gaussian-processes-128685.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.13295>)

Author: Olga Isupova, Danil Kuzin, Fredrik Gustafsson, Lyudmila Mihaylova

Published: 2026-11-11T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [sequential-testing](<https://devfeed.tech/topics/sequential-testing.md>)

Tags: [change](<https://devfeed.tech/tags/change.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [detection](<https://devfeed.tech/tags/detection.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gaussian](<https://devfeed.tech/tags/gaussian.md>), [hypothesis](<https://devfeed.tech/tags/hypothesis.md>), [hypothesis-testing](<https://devfeed.tech/tags/hypothesis-testing.md>), [math-st](<https://devfeed.tech/tags/math-st.md>), [multiple](<https://devfeed.tech/tags/multiple.md>), [offline](<https://devfeed.tech/tags/offline.md>), [online](<https://devfeed.tech/tags/online.md>), [performance](<https://devfeed.tech/tags/performance.md>), [series](<https://devfeed.tech/tags/series.md>), [specification](<https://devfeed.tech/tags/specification.md>), [stat-me](<https://devfeed.tech/tags/stat-me.md>), [stat-ml](<https://devfeed.tech/tags/stat-ml.md>), [stat-th](<https://devfeed.tech/tags/stat-th.md>), [synthetic](<https://devfeed.tech/tags/synthetic.md>), [theoretical](<https://devfeed.tech/tags/theoretical.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

The paper proposes a Gaussian process framework that detects time series changes as alterations in the prior's hyperparameters, which are learned from data. Framed as statistical hypothesis testing, it is demonstrated for offline and online detection of multiple change points. The abstract says the method does not require detailed specifications of the time series function before and after a change and reports theoretical guarantees and evaluations on synthetic and real data.

### Source excerpt

arXiv:2611.13295v1 Announce Type: cross Abstract: This paper proposes a novel general Gaussian process framework for change point detection in time series. A change is defined as an alteration of the hyperparameters of the Gaussian process prior. We formulate the problem in the context of statistical hypothesis testing and demonstrate its performance over both offline and online multiple change points detection. The approach provides theoretical guaranties of its performance. An important advantage of the developed framework is that it does not need a detailed specification of the time series function before and after a change. The hyperparameter values are learned from the data. Comprehensive evaluation of the proposed framework is provided both on synthetic and real data and its efficiency is demonstrated.

## ERP Event Logs: A Canonical Form for KPI Time-Series Extraction and Characterization

DevFeed: [ERP Event Logs: A Canonical Form for KPI Time-Series Extraction and Characterization](<https://devfeed.tech/articles/erp-event-logs-a-canonical-form-for-kpi-time-series-extraction-and-characterization-125660.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.09903>)

Author: Sherri Hadian, Adrian Rebmann, Atacan Korkmaz, Gregor Berg, Ulf Brackmann

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>)

Tags: [arima](<https://devfeed.tech/tags/arima.md>), [characterization](<https://devfeed.tech/tags/characterization.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [customer](<https://devfeed.tech/tags/customer.md>), [erp](<https://devfeed.tech/tags/erp.md>), [ets](<https://devfeed.tech/tags/ets.md>), [event](<https://devfeed.tech/tags/event.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [industries](<https://devfeed.tech/tags/industries.md>), [kpi](<https://devfeed.tech/tags/kpi.md>), [ontology](<https://devfeed.tech/tags/ontology.md>), [order-to-cash](<https://devfeed.tech/tags/order-to-cash.md>), [procurement](<https://devfeed.tech/tags/procurement.md>), [sales](<https://devfeed.tech/tags/sales.md>), [sap](<https://devfeed.tech/tags/sap.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

A canonical ERP event log can make KPI extraction reusable across systems: after a one-time mapping, volumes, durations, and rates are computed from a shared case, activity, and timestamp structure instead of bespoke joins for each metric. The study applies this approach to order-to-cash and procure-to-pay event logs from hundreds of SAP S/4HANA customers, normalizing rates and screening for minimum coverage to aid cross-customer comparison. It reports process variation across customers and industries and a recurring forecastability ordering from volumes through durations to rates. Chronos is competitive with ARIMA and ETS on duration and rate series, while classical models retain an edge on volumes.

### Source excerpt

arXiv:2610.09903v2 Announce Type: replace Abstract: Enterprise resource planning (ERP) systems record operations as transactions spread across hundreds of normalized relational tables. Extracting an operational key performance indicator (KPI) from this schema requires a separate, bespoke join for almost every metric. We show that an event log, a canonical (case, activity, timestamp) representation derived from the same tables, collapses this heterogeneity into one flat structure, from which KPI families reduce to reusable operations after a one-time mapping. Using standardized event logs built from hundreds of SAP S/4HANA customers under a shared activity ontology, we define three families of process-derived KPIs (volumes, durations, and rates) and extract them uniformly across customer systems for two common end-to-end organizational processes, order-to-cash and procure-to-pay. We normalize rates and screen series for minimum coverage to support comparison across customers of very different sizes. We then characterize heterogeneity across industries and tenants using descriptors derived directly from the canonical form, before analyzing the resulting KPI time series. The same process executes heterogeneously across customers and industries, yet this shared representation enables direct comparison of sales- and procurement-side signals, including lead times and order arrivals. Using this shared representation, we characterize which KPI families are forecastable and find a recurring ordering from volumes through durations to rates across both processes and a highly heterogeneous customer base; a pretrained time-series foundation model (Chronos) is competitive with classical baselines such as ARIMA and ETS on rate and duration series, while classical models retain an edge on volumes.

## CHORD: Multi-resolution Compression for Multimodal Time Series

DevFeed: [CHORD: Multi-resolution Compression for Multimodal Time Series](<https://devfeed.tech/articles/chord-multi-resolution-compression-for-multimodal-time-series-126201.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.11913>)

Author: Bibhus Luitel, Payal Mohapatra, Subrata Biswas, Ekata Poudel, Sunwoong Kim, Qi Zhu, Bashima Islam

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Compression](<https://devfeed.tech/topics/compression.md>), [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>)

Tags: [cascade](<https://devfeed.tech/tags/cascade.md>), [classification](<https://devfeed.tech/tags/classification.md>), [compression](<https://devfeed.tech/tags/compression.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [forecasting](<https://devfeed.tech/tags/forecasting.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [quantization](<https://devfeed.tech/tags/quantization.md>), [resolutions](<https://devfeed.tech/tags/resolutions.md>), [time-series](<https://devfeed.tech/tags/time-series.md>), [wearable](<https://devfeed.tech/tags/wearable.md>), [wearables](<https://devfeed.tech/tags/wearables.md>)

### AI overview

CHORD encodes multimodal time series as a coarse-to-fine residual cascade, letting different temporal resolutions be decoded from one payload. Learned shapelet-inspired atoms synthesize corrections; shared pathways provide a basis across modalities, while private pathways preserve sensor-specific patterns. Finite scalar quantization fixes bit width per modality and scale, making payload size known before encoding, and scale dropout trains each prefix as a valid reconstruction point. The abstract reports results on seven wearable-sensing and forecasting benchmarks: at the 8x target, classification accuracy averaged within 0.62 points of uncompressed data, the smallest gap among compressed methods.

### Source excerpt

arXiv:2611.11913v1 Announce Type: new Abstract: Wearables, mobile devices, and IoT platforms generate multimodal time-series data continuously and at scale, making efficient storage and transfer a primary concern. The same recordings are reused after collection by many downstream models, often at different temporal resolutions, so a compressed representation should be small, useful across resolutions, and produced at a bit rate known in advance. Prior work meets parts of this: symbolic and error-bounded compressors reconstruct accurately but fix resolution at compression time, learned codecs target one signal domain, and discrete tokenizers reconstruct without accounting for rate. We introduce CHORD, a coarse-to-fine compressor for heterogeneous multimodal time series. A residual cascade orders each payload from broad to fine temporal scales and synthesizes corrections through learned shapelet-inspired atoms. A shared pathway supplies a reusable synthesis basis across modalities, while private pathways retain capacity for sensor-specific morphology. Finite scalar quantization fixes the bit width of every modality and scale, so payload size is known before encoding, and scale dropout trains each coarse-to-fine prefix to be a valid reconstruction operating point. Across seven wearable-sensing and forecasting benchmarks, CHORD stays within 0.62 points of uncompressed classification accuracy on average at the 8x target, the smallest gap of any compressed method, with all resolutions decoded from one payload.

## Scalable Counterfactual Explanations for Time Series via Policy Learning

DevFeed: [Scalable Counterfactual Explanations for Time Series via Policy Learning](<https://devfeed.tech/articles/scalable-counterfactual-explanations-for-time-series-via-policy-learning-125435.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.11704>)

Author: Peiyu Li, Pouya Hosseinzadeh, Soukaina Filali Boubrahimi, Shah Muhammad Hamdi

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Time Series](<https://devfeed.tech/topics/time-series.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [outlier-detection](<https://devfeed.tech/topics/outlier-detection.md>)

Tags: [actor](<https://devfeed.tech/tags/actor.md>), [agent](<https://devfeed.tech/tags/agent.md>), [anomaly-detection](<https://devfeed.tech/tags/anomaly-detection.md>), [autoencoder](<https://devfeed.tech/tags/autoencoder.md>), [bandit](<https://devfeed.tech/tags/bandit.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [complex](<https://devfeed.tech/tags/complex.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [detection](<https://devfeed.tech/tags/detection.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [framework](<https://devfeed.tech/tags/framework.md>), [generation](<https://devfeed.tech/tags/generation.md>), [global](<https://devfeed.tech/tags/global.md>), [inference](<https://devfeed.tech/tags/inference.md>), [learning](<https://devfeed.tech/tags/learning.md>), [policy](<https://devfeed.tech/tags/policy.md>), [series](<https://devfeed.tech/tags/series.md>), [time](<https://devfeed.tech/tags/time.md>), [time-series](<https://devfeed.tech/tags/time-series.md>)

### AI overview

The paper presents a time-series counterfactual method that replaces per-instance iterative optimization at inference with a learned policy. An actor-critic agent is trained as a one-step contextual-bandit policy in the latent space of a pre-trained autoencoder, then generates counterfactuals in one forward pass. The authors report high validity and strong plausibility across multiple anomaly-detection metrics on 13 UCR benchmark datasets. The supplied abstract does not give metric values or further experimental details.

### Source excerpt

arXiv:2611.11704v1 Announce Type: new Abstract: Generating counterfactual (CF) explanations for time series is challenging due to complex temporal dependencies and the computational inefficiency of traditional instance-specific optimization. We propose a policy learning framework for counterfactual explanation in time series, trained as a one-step contextual-bandit policy-learning problem. By training an actor-critic agent in the latent space of a pre-trained autoencoder, our method learns a transferable policy that captures global structural regularities. Unlike existing methods that rely on iterative optimization for each new input, the learned policy enables direct generation of valid and plausible counterfactuals through a single forward pass, avoiding per-instance optimization at inference time. Experiments on 13 UCR benchmark datasets show that our framework achieves high validity and strong plausibility across multiple anomaly detection metrics. Overall, by decoupling counterfactual generation from per-instance optimization, the proposed method provides an efficient and scalable alternative to existing approaches for time series counterfactual explanations.

## When Sensors Fail: Coverage Auditing with Incomplete Sensor Records

DevFeed: [When Sensors Fail: Coverage Auditing with Incomplete Sensor Records](<https://devfeed.tech/articles/when-sensors-fail-coverage-auditing-with-incomplete-sensor-records-125359.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.11324>)

Author: Chung-I Huang, Wen-Yi Chang

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Predictive AI](<https://devfeed.tech/topics/predictive-ai.md>), [Temporal data](<https://devfeed.tech/topics/temporal-data.md>), [audit](<https://devfeed.tech/topics/audit.md>)

Tags: [air-quality](<https://devfeed.tech/tags/air-quality.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [audits](<https://devfeed.tech/tags/audits.md>), [conditional](<https://devfeed.tech/tags/conditional.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [forecast](<https://devfeed.tech/tags/forecast.md>), [measurement](<https://devfeed.tech/tags/measurement.md>), [metr](<https://devfeed.tech/tags/metr.md>), [outage](<https://devfeed.tech/tags/outage.md>), [prediction](<https://devfeed.tech/tags/prediction.md>), [sensor](<https://devfeed.tech/tags/sensor.md>), [stat-ap](<https://devfeed.tech/tags/stat-ap.md>), [traffic](<https://devfeed.tech/tags/traffic.md>)

### AI overview

The audit evaluates prediction-interval coverage during sensor outages while accounting for forecasts whose future targets were never observed. It preserves complete forecast keys, groups sensors and overlapping horizons by measurement tick, and combines time-uniform bounds on covered and scored forecasts into a cumulative conditional expected coverage ratio, distinct from empirical coverage and outage-conditional population coverage. In METR-LA, only 7.6% of issued long-gap forecasts had originally observed targets. Across 30 primary sequential outage-state audits, 26 were inconclusive and four lacked scored support, limiting conclusions about long-outage coverage.

### Source excerpt

arXiv:2611.11324v1 Announce Type: new Abstract: Sensor outages can conceal prediction-interval coverage deficits behind acceptable marginal performance, while unavailable future readings leave some issued forecasts unscorable. We present a measurement-record audit that links issue-time outage duration, original target availability, and time-uniform statistical evidence. The audit retains complete forecast keys, batches all sensors and overlapping horizons at each measurement tick, and projects simultaneous covered-count and scored-count bounds onto a cumulative conditional expected coverage ratio. This target is distinguished from empirical coverage and outage-conditional population coverage. Selected traffic and air-quality results illustrate the gap between pooled and long-outage coverage; in METR-LA, only 7.6% of issued long-gap forecasts have originally observed targets. Of 30 primary sequential outage-state audits, 26 are inconclusive and four lack scored support. This concise presentation gives the core construction and selected aggregate evidence without the full experimental supplement.

[Next page](<https://devfeed.tech/topics/temporal-data.md?cursor=WyIyMDI2LTExLTEwVDA1OjAwOjAwKzAwOjAwIiwgIjhjZTU0YjhjLWZkNDEtNGE1OS1hODM2LTI2MzIwNmY5ZTM2NCJd>)