Newest papers

CausalAlign: Causal Alignment of Video Generation Models for Consistent Ego-Motion

Kai Zhu, Yintianrun Zhang, Shuwei Jiang, Ruolin Zuo, Yixin Chen, Duomin Wang, Miao Liu, Yin Li, Daquan Zhou, Yiwu Zhong · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.29915 · cited by 0 works (OpenAlex) · open-access PDF

German

Diffusion-basierte Video-Generationsmodelle (VGMs) können visuell plausible Videos produzieren, aber ihre Kamera-Ego-Bewegung fließt oft über lange Horizonte. Camera-Ego-Bewegung bestimmt, wie sich der Sichtpunkt zwischen den Frames ändert und für die Aufrechterhaltung einer konsistenten Geometrie und Trajectories unerlässlich ist. Um dieses Problem anzugehen, schlagen wir CausalAlign vor, ein Kausal-Bewegungs-Framework mit zwei Komponenten. Erstens diagnostizieren wir, wie präzise Ego-Bewegung durch den Delta-Token-Raum in VGMs dargestellt, verwendet und beibehalten wird, wobei lineare Sonden und kausale Interventionen verwendet werden, wodurch ein Aufstieg-und-

Machine translation (Meta M2M100); the English original follows.

Abstract

Diffusion-based video generation models (VGMs) can produce visually plausible videos, yet their camera ego-motion often drifts over long horizons. Camera ego-motion determines how the viewpoint changes between frames and is essential for maintaining consistent geometry and trajectories. To address this problem, we propose CausalAlign, a causal alignment framework with two components. First, we diagnose how precise ego motion is represented, used, and retained within VGMs using linear probes and causal interventions, revealing a rise-and-fall pattern during denoising. Second, we strengthen ego-motion state in VGM by aligning the representations at the identified layers to a 3D foundation model through delta-token space. This space encodes the changes between consecutive frames, allowing the alignment to focus on ego-motion dynamics rather than static scene content. With Wan2.2 fine-tuned on DL3DV and RealEstate10K, CausalAlign can largely improve motion accuracy, geometric consistency, and temporal smoothness, while preserving visual quality. For example, it improves geometric consistency by more than 20% relative to the base model. Moreover, with the fine-tuned VGM, CausalAlign improves the success rate of world action models (WAMs) by 10 points on mobile manipulation tasks of RoboCasa365 benchmark. Overall, CausalAlign improves ego-motion consistency in diffusion-based VGMs and yields noticeable gains in embodied models.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer vision, Artificial intelligence, Consistency (knowledge bases), Computer science, Motion (physics), Focus (optics)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange