Kairos: Toward Adaptive and Parameter-Efficient Time Series Foundation Models

arXiv:2509.25826 · cs.LG · Submitted 2025-09-30 · Read on arXiv

cs.LG

Submitted: 2025-09-30

Updated: 2026-09-28

Comments: Accepted at NeurIPS 2026

Project page: https://foundation-model-research.github.io/Kairos

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Tokenization — 3× in this paper · explained in 2 episodes
• Foundation Model — 2× in this paper · explained in 3 episodes
• Foundation Models — 2× in this paper · explained in 7 episodes
• Abstraction — 1× in this paper · explained in 1 episode
• Adaptation — 1× in this paper · explained in 1 episode
• Benchmark — 1× in this paper · explained in 2 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Temporal Abstraction — 1× in this paper · explained in 2 episodes
• Time Series Foundation Models (TSFMs) — 1× in this paper · explained in 2 episodes
• Zero-Shot Generalization — 1× in this paper · explained in 5 episodes

Sources

• GPT-4 Technical Report
• Cosmos World Foundation Model Platform for Physical AI
• TiRex: Zero-Shot Forecasting Across Long and Short Horizons with Enhanced In-Context Learning
• Pathformer: Multi-scale Transformers with Adaptive Pathways for Time Series Forecasting
• This Time is Different: An Observability Perspective on Time Series Foundation Models
• FlowState: Sampling-Rate-Equivariant Time-Series Forecasting
• Segment Anything
• DeepSeek-V3 Technical Report
• Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Experts
• iTransformer: Inverted Transformers Are Effective for Time Series Forecasting
• Timer-XL: Long-Context Transformers for Unified Time Series Forecasting
• Sundial: A Family of Highly Capable Time Series Foundation Models
• Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts
• TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting
• Output Scaling: YingLong-Delayed Chain of Thought in a Large Pretrained Time Series Forecasting Model
• Deep Time Series Models: A Comprehensive Survey and Benchmark
• TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis
• Are Transformers Effective for Time Series Forecasting?
• ElasTST: Towards Robust Varied-Horizon Forecasting with Elastic Time-Series Transformer

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks