Reasoning-Aware Training for Time Series Forecasting

TL;DR

STRIDE is a novel framework that enhances time series forecasting models by integrating reasoning capabilities directly into their continuous embedding space, leading to state-of-the-art accuracy and interpretability.

Contribution

The paper introduces STRIDE, a method that injects reasoning traces into TSFMs via distilled embeddings, improving forecasting accuracy and reasoning interpretability.

Findings

STRIDE achieves state-of-the-art results on GIFT-Eval with 0.674 MASE.

STRIDE outperforms existing TSFMs on TFRBench in both in-domain and out-of-domain tasks.

It consistently improves diverse TSFMs like Chronos-2 and Timer-S1 across various LLM configurations.

Abstract

Time Series Foundation Models (TSFMs) excel at numerical forecasting but operate as black boxes lacking qualitative reasoning. Conversely, applying LLMs directly to temporal data introduces a modality gap: text tokenizers fragment continuous numerical values, degrading mathematical relationships and exploding sequence lengths, leading to computational overhead. To resolve this, we introduce STRIDE (Strategic Time-series Reasoning Injected via Distilled Embeddings), a novel framework natively integrating LLM reasoning into the continuous embedding space of TSFMs. Instead of discrete tokens, STRIDE distills reasoning traces into a lightweight LLM, dynamically projecting its mean-pooled hidden states as a cross-modal prior into the target numerical encoder. The architecture is jointly optimized using cross-entropy and quantile losses. Evaluations demonstrate STRIDE establishes state-of-the-art numerical forecasting on GIFT-Eval (0.674 MASE, 0.454 CRPS) compared to TSFMs and exhibits superior in-domain and out-of-domain numerical as well as reasoning performance on TFRBench. Specifically, STRIDE acts as a plug-and-play enhancement, consistently improving diverse TSFMs (e.g., Chronos-2, Timer-S1) across various LLM configurations. Thus, injecting semantic reasoning as a continuous prior equips TSFMs with human-interpretable reasoning while fundamentally improving predictive accuracy.