{"ID":6029845,"CreatedAt":"2026-07-08T02:57:47.77373338Z","UpdatedAt":"2026-07-10T17:09:12.807930246Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.06516","arxiv_id":"2607.06516","title":"Point as Skeleton: Accumulated Point Cloud Enhanced Autoregressive Generation for Closed-Loop Autonomous Driving Simulation","abstract":"Evaluating end-to-end autonomous driving (E2E-AD) remains challenging, as existing driving simulation methods often trade off closed-loop interactivity (e.g., CARLA) and real-world visual fidelity (e.g., nuScenes). We present \\textbf{\\emph{Point as Skeleton}}, a generative sensor simulation framework for state-updated autoregressive driving video generation, in which an autoregressive generator synthesizes visual observations from step-wise updated ego states, actor states, scene maps, and point-cloud skeleton conditions. To support closed-loop rollout, we introduce Reset-and-Roll, which adapts rolling diffusion inference to simulation by preventing future-conditioned latent states from being committed across simulation steps. To stabilize error accumulation during step-wise autoregressive rollout, we introduce point-cloud skeletons that decouple foreground and background assets and project them into camera-view painted-point and template-depth conditions, providing appearance and geometric cues. We further implement a nuPlan-based renderer-level closed-loop generative interface for evaluating generation under ego deviations from the original log. Experiments on nuScenes and nuPlan show that \\textit{Point as Skeleton} improves autoregressive generation quality during closed-loop rollout, demonstrating its potential for visually faithful closed-loop driving simulation. The code is available at https://github.com/krauwu/point-as-skeleton.","short_abstract":"Evaluating end-to-end autonomous driving (E2E-AD) remains challenging, as existing driving simulation methods often trade off closed-loop interactivity (e.g., CARLA) and real-world visual fidelity (e.g., nuScenes). We present \\textbf{\\emph{Point as Skeleton}}, a generative sensor simulation framework for state-updated...","url_abs":"https://arxiv.org/abs/2607.06516","url_pdf":"https://arxiv.org/pdf/2607.06516v1","authors":"[\"Songbur Wong\",\"Xiaosong Jia\",\"Junqi You\",\"Bo Zhang\",\"Pei Xu\",\"Renqiu Xia\",\"Yuping Qiu\",\"Shaofeng Zhang\",\"Zelin Zhao\",\"Xuechao Yan\",\"Yuchen Zhou\",\"Yurui Chen\",\"Wen Guo\",\"Hang Xu\",\"Junchi Yan\"]","published":"2026-07-07T17:20:51Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":614037,"CreatedAt":"2026-07-08T02:57:47.77373338Z","UpdatedAt":"2026-07-08T02:57:47.77373338Z","DeletedAt":null,"paper_id":6029845,"paper_url":"https://arxiv.org/abs/2607.06516","paper_title":"Point as Skeleton: Accumulated Point Cloud Enhanced Autoregressive Generation for Closed-Loop Autonomous Driving Simulation","repo_url":"https://github.com/krauwu/point-as-skeleton","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}