FLAME: Flow Enhanced Legendre Memory Models for General Time Series Forecasting
Not on the map. Why?
Abstract (arXiv)
In this work, we introduce FLAME, a family of extremely lightweight and capable Time Series Foundation Models, which support versatile forecasting tasks via generative probabilistic modeling, while ensuring both efficiency and robustness. FLAME utilizes the Legendre Memory for strong generalization capabilities. By adapting variants of Legendre Memory, i.e., translated Legendre (LegT) and scaled Legendre (LegS), in the Encoding and Decoding phases, FLAME can effectively capture the inherent inductive bias within data and make efficient long-range inferences. To enhance the accuracy of probabilistic forecasting while remaining efficient, FLAME adopts a normalizing-flow-based forecasting head, which can model complex distributions over the forecasting horizon in a generative manner. Comprehensive experiments on three well-recognized benchmarks, including TSFM-Bench, ProbTS, and TFB, demonstrate that FLAME is a strong out-of-the-box tool for decision intelligence.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 79 of its 115 references that are in the map. First posted on arXiv 16 Dec 2025.
Builds on
79 of 115 references.
References · 79
- 2017Attention Is All You Needcited by 59,932
- 2020Denoising Diffusion Probabilistic Modelscited by 20,851
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
- 2023GPT-4 Technical Reportcited by 18,132
- 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
- 2022Scalable Diffusion Models with Transformerscited by 5,931
- 2022Flow Matching for Generative Modelingcited by 5,426
- 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
- 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
- 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
- 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
- 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,283
38 more
- 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,187
- 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,030
- 2021Transformers in Vision: A Surveycited by 1,009
- 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 906
- 2024Chronos: Learning the Language of Time Seriescited by 712
- 2017Masked Autoregressive Flow for Density Estimationcited by 680
- 2023Time-LLM: Time Series Forecasting by Reprogramming Large Language Modelscited by 680
- 2023A decoder-only foundation model for time-series forecastingcited by 650
- 2023One Fits All:Power General Time Series Analysis by Pretrained LMcited by 615
- 2020HiPPO: Recurrent Memory with Optimal Polynomial Projectionscited by 600
- 2019Neural Spline Flowscited by 590
- 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
- 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputationcited by 577
- 2018The UCR Time Series Archivecited by 421
- 2021SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interactioncited by 335
- 2021Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecastingcited by 332
- 2018The UEA multivariate time series classification archive, 2018cited by 330
- 2023Follow Your Pose: Pose-Guided Text-to-Video Generation using Pose-Free Videoscited by 312
- 2024MACE: Mass Concept Erasure in Diffusion Modelscited by 298
- 2024Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Expertscited by 280
- 2021Monash Time Series Forecasting Archivecited by 256
- 2024TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methodscited by 242
- 2024Timer: Generative Pre-trained Transformers Are Large Time Series Modelscited by 198
- 2023TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Compositioncited by 182
- 2024Follow-Your-Emoji: Fine-Controllable and Expressive Freestyle Portrait Animationcited by 172
- 2025Sundial: A Family of Highly Capable Time Series Foundation Modelscited by 168
- 2022FiLM: Frequency improved Legendre Memory Model for Long-term Time Series Forecastingcited by 147
- 2020Multivariate Probabilistic Time Series Forecasting via Conditioned Normalizing Flowscited by 142
- 2024Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Seriescited by 120
- 2024CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-Tuningcited by 105
- 2024Robust Watermarking Using Generative Priors Against Image Editing: From Benchmarking to Advancescited by 105
- 2024DUET: Dual Clustering Enhanced Multivariate Time Series Forecastingcited by 104
- 2024Pathformer: Multi-scale Transformers with Adaptive Pathways for Time Series Forecastingcited by 102
- 2025Controllable Video Generation: A Surveycited by 93
- 2024CATCH: Channel-Aware multivariate Time Series Anomaly Detection via Frequency Patchingcited by 91
- 2023Generative Time Series Forecasting with Diffusion, Denoise, and Disentanglementcited by 89
- 2024VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasterscited by 87
- 2023Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series Forecastingcited by 86
50 most cited of 79 All 79 (JSON)
Built on by
Citing papers per year: 2026: 3.
3 papers.