TSFMAudit detects pretraining contamination in time series foundation models via probe adaptation dynamics (faster loss drop, smaller backbone shift), tested on 6 models and 187 datasets against 10 LLM-derived baselines.
hub
Tirex: Zero-shot forecasting across long and short horizons with enhanced in-context learning
23 Pith papers cite this work. Polarity classification is still indexing.
23
Pith papers citing it
hub tools
citation-role summary
baseline 4
citation-polarity summary
roles
baseline 3polarities
baseline 3representative citing papers
Chronicle is the first model jointly pretrained from scratch on text and time series in a unified transformer that matches a comparable language model on NLU tasks and sets new bars for time series classification and multimodal forecasting.
A model-agnostic adaptive conformal anomaly detection approach uses weighted quantile bounds learned from past foundation model predictions to deliver interpretable p-value scores with stable calibration under shifts for time series monitoring.
TempusBench is a new evaluation framework for time-series forecasting models that supplies fresh non-overlapping datasets, tasks beyond horizon and domain, consistent tuning across models, and visualization tools.
Flow matching on time series targets a closed-form nonparametric velocity field that is a similarity-weighted mixture of observed transition velocities, making neural models approximations to an ideal memory-augmented dynamical system sampler.
TS-Arena is a live pre-registration platform that evaluates time series forecasts on future data streams to eliminate information leakage.
Compares foundation models for probabilistic low-voltage load forecasting on 200 real feeders and introduces a grid-planning metric that scores peak prediction by its effect on asset cost-risk decisions.
TimeRouter routes among time-series foundation models via discriminative routing, selective gating and ensemble fallback, reporting SOTA LB MASE 0.6765 on GIFT-EVAL.
Tyan-WP is a pretrained wind power foundation model that outperforms site-specific TSMs and generic LTSMs in zero-shot ultra-short-term probabilistic forecasting on U.S. and U.K. sites via static embeddings and PAMF module.
A zero-shot pipeline with physics-informed synthetic histories lets time-series foundation models outperform baselines by 1.7-2x in cold-start PV forecasting on 440 sites across four datasets.
GITCO delivers +1.95% average MASE reduction on TimesFM 2.5 across 53 datasets by gated inference-time suppression of anomalous patches, capturing 89.9% of the improvement upper bound.
AME-TS is a structure-guided sparse MoE foundation model for time series that aligns expert routing with series-level temporal descriptors to achieve strong accuracy-efficiency tradeoffs on GIFT-Eval while improving specialization stability.
TabPFN-3 scales tabular foundation models to 1M rows with synthetic pretraining, test-time compute, and benchmark-leading performance on tabular, relational, and tabular-text tasks while being up to 20x faster than TabPFN-2.5.
RareCP improves interval efficiency for time series conformal prediction by retrieving and weighting regime-specific calibration examples while adapting to drift and maintaining coverage.
Foundation models outperform dataset-specific machine learning in energy time series forecasting across 54 datasets in 9 categories.
iAmTime is a time-series foundation model that uses instruction-conditioned in-context learning from demonstrations to perform zero-shot adaptation on forecasting, imputation, classification, and related tasks.
Timer-S1 is a released 8.3B-parameter MoE time series model that achieves state-of-the-art MASE and CRPS scores on GIFT-Eval using serial scaling and Serial-Token Prediction.
X-TRACK is the first xLSTM model with explicit kinematic constraints that generates realistic highway trajectories and outperforms baselines on highD while matching SOTA on NGSIM.
Kairos is a parameter-efficient time series foundation model using dynamic patching tokenizer, mixture-of-size encoding, and spectral-conditioned positional embeddings to improve zero-shot forecasting on heterogeneous data.
NormWear-2 encodes physiological signals and interventions into a shared latent space, models their joint evolution as a dynamical system, and uses chaos-theoretic balancing during pretraining to achieve superior multi-scale forecasting on diverse real-world datasets.
Embedding spaces of time series foundation models make mean shifts, variance changes, and trends linearly detectable, but detection degrades smoothly with shift strength and shows model-specific failure modes.
Normalization choice significantly influences training convergence and forecasting performance in causal large time-series models.
AI offers opportunities to advance fusion energy R&D but requires responsible practices and expert collaborations to overcome its inherent challenges.
citing papers explorer
-
TSFMAudit: Data Contamination Auditing in Forecasting Time Series Foundation Models
TSFMAudit detects pretraining contamination in time series foundation models via probe adaptation dynamics (faster loss drop, smaller backbone shift), tested on 6 models and 187 datasets against 10 LLM-derived baselines.
-
Chronicle: A Multimodal Foundation Model for Joint Language and Time Series Understanding
Chronicle is the first model jointly pretrained from scratch on text and time series in a unified transformer that matches a comparable language model on NLU tasks and sets new bars for time series classification and multimodal forecasting.
-
Adaptive Conformal Anomaly Detection with Time Series Foundation Models for Signal Monitoring
A model-agnostic adaptive conformal anomaly detection approach uses weighted quantile bounds learned from past foundation model predictions to deliver interpretable p-value scores with stable calibration under shifts for time series monitoring.
-
TempusBench: An Evaluation Framework for Time-Series Forecasting
TempusBench is a new evaluation framework for time-series forecasting models that supplies fresh non-overlapping datasets, tasks beyond horizon and domain, consistent tuning across models, and visualization tools.
-
Is Flow Matching Just Trajectory Replay for Sequential Data?
Flow matching on time series targets a closed-form nonparametric velocity field that is a similarity-weighted mixture of observed transition velocities, making neural models approximations to an ideal memory-augmented dynamical system sampler.
-
TS-Arena -- A Live Forecast Pre-Registration Platform
TS-Arena is a live pre-registration platform that evaluates time series forecasts on future data streams to eliminate information leakage.
-
Probabilistic Low-Voltage Peak Load Forecasting with Time Series Foundation Models Evaluated on Application-Oriented Metrics
Compares foundation models for probabilistic low-voltage load forecasting on 200 real feeders and introduces a grid-planning metric that scores peak prediction by its effect on asset cost-risk decisions.
-
TimeRouter: Efficient and Adaptive Routing of Time-Series Foundation Models
TimeRouter routes among time-series foundation models via discriminative routing, selective gating and ensemble fallback, reporting SOTA LB MASE 0.6765 on GIFT-EVAL.
-
Tyan-WP: A Wind Power Foundation Model for Ultra-Short-Term Probabilistic Forecasting
Tyan-WP is a pretrained wind power foundation model that outperforms site-specific TSMs and generic LTSMs in zero-shot ultra-short-term probabilistic forecasting on U.S. and U.K. sites via static embeddings and PAMF module.
-
Time series Foundation Models based on Physics-Informed Synthetic Histories for Cold-Start Photovoltaic Forecasting
A zero-shot pipeline with physics-informed synthetic histories lets time-series foundation models outperform baselines by 1.7-2x in cold-start PV forecasting on 440 sites across four datasets.
-
GITCO: Gated Inference-Time Context Optimization in TSFMs
GITCO delivers +1.95% average MASE reduction on TimesFM 2.5 across 53 datasets by gated inference-time suppression of anomalous patches, capturing 89.9% of the improvement upper bound.
-
AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting
AME-TS is a structure-guided sparse MoE foundation model for time series that aligns expert routing with series-level temporal descriptors to achieve strong accuracy-efficiency tradeoffs on GIFT-Eval while improving specialization stability.
-
TabPFN-3: Technical Report
TabPFN-3 scales tabular foundation models to 1M rows with synthetic pretraining, test-time compute, and benchmark-leading performance on tabular, relational, and tabular-text tasks while being up to 20x faster than TabPFN-2.5.
-
RareCP: Regime-Aware Retrieval for Efficient Conformal Prediction
RareCP improves interval efficiency for time series conformal prediction by retrieving and weighting regime-specific calibration examples while adapting to drift and maintaining coverage.
-
FETS Benchmark: Foundation Models Outperform Dataset-specific Machine Learning in Energy Time Series Forecasting
Foundation models outperform dataset-specific machine learning in energy time series forecasting across 54 datasets in 9 categories.
-
A Foundation Model for Instruction-Conditioned In-Context Time Series Tasks
iAmTime is a time-series foundation model that uses instruction-conditioned in-context learning from demonstrations to perform zero-shot adaptation on forecasting, imputation, classification, and related tasks.
-
Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling
Timer-S1 is a released 8.3B-parameter MoE time series model that achieves state-of-the-art MASE and CRPS scores on GIFT-Eval using serial scaling and Serial-Token Prediction.
-
X-TRACK: Physics-Aware xLSTM for Realistic Vehicle Trajectory Prediction
X-TRACK is the first xLSTM model with explicit kinematic constraints that generates realistic highway trajectories and outperforms baselines on highD while matching SOTA on NGSIM.
-
Kairos: Toward Adaptive and Parameter-Efficient Time Series Foundation Models
Kairos is a parameter-efficient time series foundation model using dynamic patching tokenizer, mixture-of-size encoding, and spectral-conditioned positional embeddings to improve zero-shot forecasting on heterogeneous data.
-
Toward World Modeling of Physiological Signals with Chaos-Theoretic Balancing and Latent Dynamics
NormWear-2 encodes physiological signals and interventions into a shared latent space, models their joint evolution as a dynamical system, and uses chaos-theoretic balancing during pretraining to achieve superior multi-scale forecasting on diverse real-world datasets.
-
Non-Stationarity in the Embedding Space of Time Series Foundation Models
Embedding spaces of time series foundation models make mean shifts, variance changes, and trends linearly detectable, but detection degrades smoothly with shift strength and shows model-specific failure modes.
-
Does Normalization Choice Matter for Causal Large Time-Series Models?
Normalization choice significantly influences training convergence and forecasting performance in causal large time-series models.
-
Challenges and opportunities for AI to help deliver fusion energy
AI offers opportunities to advance fusion energy R&D but requires responsible practices and expert collaborations to overcome its inherent challenges.