[
    {
        "id": "osp-24094",
        "type": "article-journal",
        "title": "VASC: Value-Aware Sparse Attention with Cross-Layer Memory for Efficient 3D Reconstruction",
        "author": [
            {
                "family": "Wu",
                "given": "Junyi"
            },
            {
                "family": "Kong",
                "given": "Fanqing"
            },
            {
                "family": "Chen",
                "given": "Leyang"
            },
            {
                "family": "Zhang",
                "given": "Shaoqiu"
            },
            {
                "family": "Zhang",
                "given": "Yulun"
            }
        ],
        "URL": "https://omanscience.com/en/articles/vasc-value-aware-sparse-attention-with-cross-layer-memory-for-efficient-3d-reconstruction",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Feed-forward 3D vision models such as VGGT have achieved remarkable progress, unifying camera estimation and dense scene reconstruction in a single pass. However, their quadratic global attention makes long image sequences expensive, while existing sparse methods may favor highly attended yet value-redundant regions. To address these limitations, we introduce VASC, a training-free sparse attention method combining value-aware block selection and execution-aware cross-layer memory. Our value-aware block selection integrates pooled query--key relevance with neighboring value contrast, reducing redundancy while preserving query-relevant and distinctive content. Cross-layer memory tracks unserved demand across layers and updates this state according to actual execution, enabling previously underserved blocks to compete under a fixed computation budget. Experiments on 7Scenes and NeuralRGB-D with VGGT and $π^3$ demonstrate improved pose estimation and reconstruction quality compared with FasterVGGT, together with up to $2.29\\times$ faster inference than dense VGGT. Code is available at https://github.com/kosakayamahoo-design/VASC."
    }
]