{"ID":2876971,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.20376","arxiv_id":"2508.20376","title":"Enhancing Mamba Decoder with Bidirectional Interaction in Multi-Task Dense Prediction","abstract":"Sufficient cross-task interaction is crucial for success in multi-task dense prediction. However, sufficient interaction often results in high computational complexity, forcing existing methods to face the trade-off between interaction completeness and computational efficiency. To address this limitation, this work proposes a Bidirectional Interaction Mamba (BIM), which incorporates novel scanning mechanisms to adapt the Mamba modeling approach for multi-task dense prediction. On the one hand, we introduce a novel Bidirectional Interaction Scan (BI-Scan) mechanism, which constructs task-specific representations as bidirectional sequences during interaction. By integrating task-first and position-first scanning modes within a unified linear complexity architecture, BI-Scan efficiently preserves critical cross-task information. On the other hand, we employ a Multi-Scale Scan~(MS-Scan) mechanism to achieve multi-granularity scene modeling. This design not only meets the diverse granularity requirements of various tasks but also enhances nuanced cross-task feature interactions. Extensive experiments on two challenging benchmarks, \\emph{i.e.}, NYUD-V2 and PASCAL-Context, show the superiority of our BIM vs its state-of-the-art competitors.","short_abstract":"Sufficient cross-task interaction is crucial for success in multi-task dense prediction. However, sufficient interaction often results in high computational complexity, forcing existing methods to face the trade-off between interaction completeness and computational efficiency. To address this limitation, this work pro...","url_abs":"https://arxiv.org/abs/2508.20376","url_pdf":"https://arxiv.org/pdf/2508.20376v1","authors":"[\"Mang Cao\",\"Sanping Zhou\",\"Yizhe Li\",\"Ye Deng\",\"Wenli Huang\",\"Le Wang\"]","published":"2025-08-28T02:50:19Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}