NeurIPS 2026 accepted papers on Time series
122 of the 6,080 papers accepted at NeurIPS 2026 (2.0%) are on Time series, compared with 1.7% at NeurIPS 2025. They fall into 4 topics.
See NeurIPS 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Causal and generative time series · 49 papers
- RamanBench: A Large-Scale Benchmark for Machine Learning on Raman Spectroscopy (1 citation)
- Calibrating Scientific Foundation Models with Inference-Time Stochastic Attention
- Curriculum Multiple Shooting for Robust Training of Neural and Universal Differential Equations
- DiffPTS: Rethinking Diffusion ELBO for Probabilistic Time Series Forecasting
- Discretizing Continuous Time Series for Imputation with Masked Diffusion Training
- E4GEN: Event-level Explainable Extreme-Enhanced Time-series Generation
- Faithful Embeddings of Irregular and Asynchronous Data for Online Log-NCDEs
- Filtered Conformal Ellipsoids for Graph-Native Time Series
- Generating Financial Time Series by Matching Random Convolutional Features
- High-dimensional Gaussian Graphical Model Testing for Long-Memory Time Series
- HyFAD: Hybrid Time-Frequency Diffusion with Frequency-Aware Embedding for Time Series Imputation
- Hyper Hawkes Processes: Interpretable Models of Marked Temporal Point Processes
and 37 more in the interactive map.
Foundation models for time series · 33 papers
- OTIS: Learning High-Quality Time Series Features With Tiny Encoders (2 citations)
- Align-RAG: Alignment Is All You Need for TSFM In-Context Learning
- Dynamic Tokenization via Reinforcement Patching: End-to-end Training and Zero-shot Transfer
- Empowering Time Series Analysis with Large-Scale Multimodal Pretraining
- GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting
- Learning the Context of Errors: Black-Box Online Adaptation of Time Series Foundation Models
- LLM Flow Processes for Text-Conditioned Regression
- LLMs can construct powerful representations and streamline sample-efficient supervised learning
- NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces
- OATS: Online Data Augmentation for Time Series Foundation Models
- OpenMHC: Accelerating the Science of Wearable Foundation Models
- Rethinking Post-Training Recipes for Multimodal Time-Series Forecasting
and 21 more in the interactive map.
Long-term forecasting · 27 papers
- A Set-Sequence Model for Time Series
- Dirichlet-Guided Group Forecasting for Alleviating Over-smoothing in Time Series Forecasting
- Does Synthetic Data Help? Empirical Evidence from Deep Learning Time Series Forecasters
- LogSig-SSM: Time-Series Modelling with Multi-Scale Log-Signature Compression for State-Space Models
- Multivariate Time Series Forecasting needs Cross Variable Loss
- Spectral Transformer Neural Processes
- Beyond Isolation: Neighbor-Consistent Data Pruning for Multivariate Time Series Forecasting
- BucpTSF: Breaking the Uniform Computation Paradigm in Time-Series Forecasting
- DSR-TSF: Spectrum-Driven Dynamic Routing for Efficient Long-Horizon Time Series Forecasting
- Entropy Guided Dynamic Patch Segmentation for Time Series Transformers
- FISC: Time-Series Forecasting via First-Layer Statistical Calibration Constraints
- MSAR: Next-Scale Autoregressive Forecasting for Time Series via Modular Multi-Scale Decoupling
and 15 more in the interactive map.
Anomaly detection · 13 papers
- A Locally Tokenized Generative Model for Robust Time-Series Watermarking
- COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detection
- Detect, Explain, Interpret: An End-to-End Benchmark for Time Series Anomaly Detection, Explainability and Interpretability.
- STAR: Boosting Time Series Foundation Models for Anomaly Detection Through State-Aware Adapter
- TraXion: Rethinking Pre-training Frameworks for Mobility and Beyond
- VACE: Learning Geometrically Structured Representations for Time Series Anomaly Detection
- VETime: Vision Enhanced Zero-Shot Time Series Anomaly Detection
- dStructAD: Domain-Level Structured Normality Representation with Variation Calibration for Time Series Anomaly Detection
- FENet: Functional Embedding Neural Network for Change-Point Detection in Functional Time Series
- GAPS: Gradient-Aware Adaptation-Gap Scoring for Time-Series Anomaly Detection with Foundation Models
- Retrieval Over Training: Similarity search-based Model Selection for Time Series Anomaly Detection
- Spikes as Detectors: Phase-Conditioned Spiking Dynamics for Time-Series Anomaly Detection
and 1 more in the interactive map.