Research

Paper

AI LLM March 13, 2026

From Text to Forecasts: Bridging Modality Gap with Temporal Evolution Semantic Space

Authors

Lehui Li, Yuyao Wang, Jisheng Yan, Wei Zhang, Jinliang Deng, Haoliang Sun, Zhongyi Han, Yongshun Gong

Abstract

Incorporating textual information into time-series forecasting holds promise for addressing event-driven non-stationarity; however, a fundamental modality gap hinders effective fusion: textual descriptions express temporal impacts implicitly and qualitatively, whereas forecasting models rely on explicit and quantitative signals. Through controlled semi-synthetic experiments, we show that existing methods over-attend to redundant tokens and struggle to reliably translate textual semantics into usable numerical cues. To bridge this gap, we propose TESS, which introduces a Temporal Evolution Semantic Space as an intermediate bottleneck between modalities. This space consists of interpretable, numerically grounded temporal primitives (mean shift, volatility, shape, and lag) extracted from text by an LLM via structured prompting and filtered through confidence-aware gating. Experiments on four real-world datasets demonstrate up to a 29 percent reduction in forecasting error compared to state-of-the-art unimodal and multimodal baselines. The code will be released after acceptance.

Metadata

arXiv ID: 2603.12664
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-13
Fetched: 2026-03-16 06:01

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.12664v1</id>\n    <title>From Text to Forecasts: Bridging Modality Gap with Temporal Evolution Semantic Space</title>\n    <updated>2026-03-13T05:11:54Z</updated>\n    <link href='https://arxiv.org/abs/2603.12664v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.12664v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Incorporating textual information into time-series forecasting holds promise for addressing event-driven non-stationarity; however, a fundamental modality gap hinders effective fusion: textual descriptions express temporal impacts implicitly and qualitatively, whereas forecasting models rely on explicit and quantitative signals. Through controlled semi-synthetic experiments, we show that existing methods over-attend to redundant tokens and struggle to reliably translate textual semantics into usable numerical cues. To bridge this gap, we propose TESS, which introduces a Temporal Evolution Semantic Space as an intermediate bottleneck between modalities. This space consists of interpretable, numerically grounded temporal primitives (mean shift, volatility, shape, and lag) extracted from text by an LLM via structured prompting and filtered through confidence-aware gating. Experiments on four real-world datasets demonstrate up to a 29 percent reduction in forecasting error compared to state-of-the-art unimodal and multimodal baselines. The code will be released after acceptance.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-13T05:11:54Z</published>\n    <arxiv:comment>15 pages, 6 figures</arxiv:comment>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Lehui Li</name>\n    </author>\n    <author>\n      <name>Yuyao Wang</name>\n    </author>\n    <author>\n      <name>Jisheng Yan</name>\n    </author>\n    <author>\n      <name>Wei Zhang</name>\n    </author>\n    <author>\n      <name>Jinliang Deng</name>\n    </author>\n    <author>\n      <name>Haoliang Sun</name>\n    </author>\n    <author>\n      <name>Zhongyi Han</name>\n    </author>\n    <author>\n      <name>Yongshun Gong</name>\n    </author>\n  </entry>"
}