{"ID":2896306,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.08153","arxiv_id":"2507.08153","title":"ALCo-FM: Adaptive Long-Context Foundation Model for Accident Prediction","abstract":"Traffic accidents are rare, yet high-impact events that require long-context multimodal reasoning for accurate risk forecasting. In this paper, we introduce ALCo-FM, a unified adaptive long-context foundation model that computes a volatility pre-score to dynamically select context windows for input data and encodes and fuses these multimodal data via shallow cross attention. Following a local GAT layer and a BigBird-style sparse global transformer over H3 hexagonal grids, coupled with Monte Carlo dropout for confidence, the model yields superior, well-calibrated predictions. Trained on data from 15 US cities with a class-weighted loss to counter label imbalance, and fine-tuned with minimal data on held-out cities, ALCo-FM achieves 0.94 accuracy, 0.92 F1, and an ECE of 0.04, outperforming more than 20 state-of-the-art baselines in large-scale urban risk prediction. Code and dataset are available at: https://github.com/PinakiPrasad12/ALCo-FM","short_abstract":"Traffic accidents are rare, yet high-impact events that require long-context multimodal reasoning for accurate risk forecasting. In this paper, we introduce ALCo-FM, a unified adaptive long-context foundation model that computes a volatility pre-score to dynamically select context windows for input data and encodes and...","url_abs":"https://arxiv.org/abs/2507.08153","url_pdf":"https://arxiv.org/pdf/2507.08153v1","authors":"[\"Pinaki Prasad Guha Neogi\",\"Ahmad Mohammadshirazi\",\"Rajiv Ramnath\"]","published":"2025-07-10T20:22:26Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Transformer\"]","has_code":false,"code_links":[{"ID":612264,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2896306,"paper_url":"https://arxiv.org/abs/2507.08153","paper_title":"ALCo-FM: Adaptive Long-Context Foundation Model for Accident Prediction","repo_url":"https://github.com/PinakiPrasad12/ALCo-FM","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}