{"ID":2835164,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.00425","arxiv_id":"2512.00425","title":"What about gravity in video generation? Post-Training Newton's Laws with Verifiable Rewards","abstract":"Recent video diffusion models can synthesize visually compelling clips, yet often violate basic physical laws-objects float, accelerations drift, and collisions behave inconsistently-revealing a persistent gap between visual realism and physical realism. We propose $\\texttt{NewtonRewards}$, the first physics-grounded post-training framework for video generation based on $\\textit{verifiable rewards}$. Instead of relying on human or VLM feedback, $\\texttt{NewtonRewards}$ extracts $\\textit{measurable proxies}$ from generated videos using frozen utility models: optical flow serves as a proxy for velocity, while high-level appearance features serve as a proxy for mass. These proxies enable explicit enforcement of Newtonian structure through two complementary rewards: a Newtonian kinematic constraint enforcing constant-acceleration dynamics, and a mass conservation reward preventing trivial, degenerate solutions. We evaluate $\\texttt{NewtonRewards}$ on five Newtonian Motion Primitives (free fall, horizontal/parabolic throw, and ramp sliding down/up) using our newly constructed large-scale benchmark, $\\texttt{NewtonBench-60K}$. Across all primitives in visual and physics metrics, $\\texttt{NewtonRewards}$ consistently improves physical plausibility, motion smoothness, and temporal coherence over prior post-training methods. It further maintains strong performance under out-of-distribution shifts in height, speed, and friction. Our results show that physics-grounded verifiable rewards offer a scalable path toward physics-aware video generation.","short_abstract":"Recent video diffusion models can synthesize visually compelling clips, yet often violate basic physical laws-objects float, accelerations drift, and collisions behave inconsistently-revealing a persistent gap between visual realism and physical realism. We propose $\\texttt{NewtonRewards}$, the first physics-grounded p...","url_abs":"https://arxiv.org/abs/2512.00425","url_pdf":"https://arxiv.org/pdf/2512.00425v1","authors":"[\"Minh-Quan Le\",\"Yuanzhi Zhu\",\"Vicky Kalogeiton\",\"Dimitris Samaras\"]","published":"2025-11-29T10:04:50Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false}