{"ID":2881290,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.13327","arxiv_id":"2508.13327","title":"Towards Unified Multimodal Financial Forecasting: Integrating Sentiment Embeddings and Market Indicators via Cross-Modal Attention","abstract":"We propose STONK (Stock Optimization using News Knowledge), a multimodal framework integrating numerical market indicators with sentiment-enriched news embeddings to improve daily stock-movement prediction. By combining numerical \u0026 textual embeddings via feature concatenation and cross-modal attention, our unified pipeline addresses limitations of isolated analyses. Backtesting shows STONK outperforms numeric-only baselines. A comprehensive evaluation of fusion strategies and model configurations offers evidence-based guidance for scalable multimodal financial forecasting. Source code is available on GitHub","short_abstract":"We propose STONK (Stock Optimization using News Knowledge), a multimodal framework integrating numerical market indicators with sentiment-enriched news embeddings to improve daily stock-movement prediction. By combining numerical \u0026 textual embeddings via feature concatenation and cross-modal attention, our unified pipe...","url_abs":"https://arxiv.org/abs/2508.13327","url_pdf":"https://arxiv.org/pdf/2508.13327v1","authors":"[\"Sarthak Khanna\",\"Armin Berger\",\"David Berghaus\",\"Tobias Deusser\",\"Lorenz Sparrenberg\",\"Rafet Sifa\"]","published":"2025-08-18T19:22:39Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[]","has_code":false}