{"ID":2881622,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.11954","arxiv_id":"2508.11954","title":"UniCast: A Unified Framework for Instance-Conditioned Multimodal Time-Series Forecasting","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.","short_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 an...","url_abs":"https://arxiv.org/abs/2508.11954","url_pdf":"https://arxiv.org/pdf/2508.11954v2","authors":"[\"Sehyuk Park\",\"Soyeon Caren Han\",\"Eduard Hovy\"]","published":"2025-08-16T07:33:27Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Transformer\"]","has_code":false}