Efficient Test-Time Scaling for LLM-based Time Series Forecasting
Xuan-May Le, Minh-Tuan Tran, Ling Luo, Uwe Aickelin, Dinh Phung, Trung Le
cs.LG
Submitted: 2026-08-09
Updated: 2026-08-11
Comments: Accepted at KDD 2026 (Oral)
Code: https://github.com/xuanmay2701/SCALER
License: http://creativecommons.org/licenses/by/4.0/
The gist: Long-term time series forecasting benefits from preserving global structure such as trends and seasonality.
Terminology
Abstract
Long-term time series forecasting benefits from preserving global structure such as trends and seasonality. Recent LLM-based forecasters often improve accuracy through test-time scaling (e.g., iterative refinement), but these methods are computationally expensive and increasingly prone to global-shape mismatch as the prediction horizon extends. We propose SCALER, a coarse-to-fine forecasting framework that first employs a lightweight Transformer tailored to long-term shape modeling to predict a coarse representation of future dynamics. This predicted shape then serves as a compact guide for an LLM to perform test-time scaling via iterative coarse-to-fine residual token refinement, while processing substantially fewer tokens at each step. By guiding refinement with an explicit future-shape prediction, SCALER reduces reliance on long description prompts, and its fixed-step refinement avoids costly reward-model-based selection, further lowering computational overhead. Experimental results demonstrate that SCALER outperforms strong forecasting baselines in long-term, short-term and zero-shot forecasting while significantly reducing the inference cost associated with scaled LLM for time series forecasting. Code: https://github.com/xuanmay2701/SCALER.
Sources
- Chronos: Learning the Language of Time Series
- Can Slow-thinking LLMs Reason Over Time? Empirical Studies in Time Series Forecasting
- Time-LLM: Time Series Forecasting by Reprogramming Large Language Models
- Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping
- TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment
- iTransformer: Inverted Transformers Are Effective for Time Series Forecasting
- TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting
- TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis
- Enhancing Large Language Models for Time-Series Forecasting via Vector-Injected In-Context Learning
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