UniCast: A Unified Framework for Instance-Conditioned Multimodal Time-Series Forecasting

TL;DR

UniCast introduces a multimodal, instance-conditioned framework for time series forecasting that enhances adaptability and accuracy by leveraging dynamic prompts and modality routing without retraining the core model.

Contribution

The paper presents UniCast, a novel parameter-efficient multimodal framework that extends TSFMs with instance-conditioned prompting and dynamic modality routing, improving forecasting performance.

Findings

Outperforms existing TSFM baselines across benchmarks.

Effectively leverages multimodal inputs for better predictions.

Maintains foundation model generalization while enabling multimodal control.

Abstract

Time series forecasting underpins applications in finance, healthcare, and environmental monitoring. Despite the success of Time Series Foundation Models (TSFMs), existing approaches operate in a unimodal setting and rely on static prompts or fixed fusion schemes, limiting their ability to exploit multimodal context and adapt to instance-level variation. We propose UniCast, a parameter-efficient multimodal framework that extends TSFMs through instance conditioned prompting and dynamic modality routing. UniCast infers a conditional prompt from time series, vision, and text inputs via a Transformer-based contextual distiller, enabling input-specific adaptation without updating the forecasting backbone. To regulate how auxiliary modalities influence predictions, UniCast employs Modality Routing, a cross-attention mechanism that estimates modality relevance given the current temporal state and selectively amplifies informative signals while suppressing noise. Integrated with a frozen TSFM via soft prompt tuning, UniCast preserves foundation-level generalization while enabling effective multimodal control. Extensive experiments across diverse forecasting benchmarks show that UniCast consistently outperforms all existing TSFM baselines, demonstrating that instance-conditioned multimodal control is critical for next-generation time series forecasting.