[
    {
        "id": "osp-17725",
        "type": "article-journal",
        "title": "IntactWorld: Joint World Modeling with Intact Features",
        "author": [
            {
                "family": "Tan",
                "given": "Boming"
            },
            {
                "family": "Zhang",
                "given": "Xiangdong"
            },
            {
                "family": "Xia",
                "given": "Yan"
            },
            {
                "family": "Zhu",
                "given": "Qi"
            },
            {
                "family": "Ji",
                "given": "Deyi"
            },
            {
                "family": "Yang",
                "given": "Xue"
            },
            {
                "family": "Zhang",
                "given": "Shaofeng"
            }
        ],
        "URL": "https://omanscience.com/en/articles/intactworld-joint-world-modeling-with-intact-features",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "While recent video generation models synthesize highly realistic visuals, they lack a genuine understanding of intrinsic real-world logic. Existing methods attempt to understand the world by internalizing diverse world knowledge, yet constrained by computational overhead or dimensionality alignment, their learning processes inevitably compress features, causing a severe loss of structural information. To address this, we propose \\textbf{IntactWorld}, a \\textbf{Joint World Modeling Architecture} utilizing uncompressed \\textbf{Intact Features}. Since data naturally reside on a low-dimensional manifold within a high-dimensional space, predicting the flow velocity $v$ within this uncompressed high-dimensional space induces a severe manifold gap. To successfully eliminate this optimization bottleneck, our framework instead predicts the clean feature $x_0$ at intermediate layers. Furthermore, to mitigate the computational overhead of incorporating complete world knowledge, we introduce a \\textit{Full-to-Compact Training Paradigm}. By replacing raw full features with highly refined CLS tokens, this paradigm enables efficient single-branch guidance, reducing spatial memory consumption by 11.4\\% and cutting inference latency by 43.8\\%. Extensive evaluations demonstrate the effectiveness of IntactWorld, outperforming established baselines by 2.46 points on the VBench 2.0 benchmark."
    }
]