{"slug": "fluid-gen-zero-grounding-pretrained-video-generators-in-physics-without-training", "title": "Fluid-Gen-Zero: Grounding Pretrained Video Generators in Physics without Training", "summary": "Fluid-Gen-Zero, a training-free framework from researchers behind the arXiv:2610.10984v1 paper, grounds pretrained video generators in physics by delegating motion dynamics to a simulator while preserving the appearance modeling of the base models. Across Tora (CogVideoX-based), VACE and WanMove (Wan-based), the framework reduced object trajectory error by 26.7%-81.5% and fluid flow endpoint error (fEPE) by 67.9%-84.0% while largely preserving perceptual quality. In a human preference study, raters favored Fluid-Gen-Zero in 55.1%-74.4% of same-backbone comparisons across three backbones and in 90.4%-94.2% of comparisons against simulation-based methods.", "body_md": "arXiv:2610.10984v1 Announce Type: new \nAbstract: We present Fluid-Gen-Zero, a training-free framework for physics-aware fluid-object interaction video generation that decouples physical reasoning from appearance synthesis. Our key insight is to delegate motion dynamics to a physics simulator while preserving the appearance modeling capacity of pretrained video generators. We bridge these two domains through a two-level agentic workflow: generation-time planning, where a vision-language model (VLM) agent interprets intent and the simulation rollout to organize generation clips, and latent-space guidance, which injects simulation signals into denoising through region-aware latent wrapping. This plug-and-play design is compatible with current video foundation models. We further introduce a benchmark for fluid-object interaction video generation. Across Tora (CogVideoX-based), VACE and WanMove (Wan-based), Fluid-Gen-Zero consistently improves simulation alignment, reducing object trajectory error by 26.7%-81.5% and fluid fEPE (fluid flow endpoint error) by 67.9%-84.0%, while largely preserving perceptual quality. In a human preference study, raters favor Fluid-Gen-Zero in 55.1%-74.4% of same-backbone comparisons across three backbones, and in 90.4%-94.2% of comparisons against simulation-based methods. Code and data will be released upon acceptance.", "url": "https://wpnews.pro/news/fluid-gen-zero-grounding-pretrained-video-generators-in-physics-without-training", "canonical_source": "https://arxiv.org/abs/2610.10984", "published_at": "2026-10-09 04:00:00+00:00", "updated_at": "2026-10-09 04:17:18.615568+00:00", "lang": "en", "topics": ["artificial-intelligence", "generative-ai", "ai-research", "computer-vision", "ai-agents"], "entities": ["Fluid-Gen-Zero", "Tora", "CogVideoX", "VACE", "WanMove", "Wan", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/fluid-gen-zero-grounding-pretrained-video-generators-in-physics-without-training", "markdown": "https://wpnews.pro/news/fluid-gen-zero-grounding-pretrained-video-generators-in-physics-without-training.md", "text": "https://wpnews.pro/news/fluid-gen-zero-grounding-pretrained-video-generators-in-physics-without-training.txt", "jsonld": "https://wpnews.pro/news/fluid-gen-zero-grounding-pretrained-video-generators-in-physics-without-training.jsonld"}}