IntactWorld: Joint World Modeling with Intact Features

arXiv:2610.11174v1 Announce Type: new 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 extbf{IntactWorld}, a extbf{Joint World Modeling Architecture} utilizing uncompressed extbf{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 extit{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.

Source: arXiv cs.CV | 2026-10-09