SparseWorld-TC：轨迹条件稀疏占用世界模型
SparseWorld-TC: Trajectory-Conditioned Sparse Occupancy World Model
用稀疏占用表示替代鸟瞰图投影与离散 VAE token，直接在原始图像特征上端到端预测多帧未来占用，并以任意未来轨迹作为条件；做自动驾驶场景预测或占用世界模型时，可借鉴其绕过 BEV 几何先验、保留连续表征的思路。
原文依据 we employ a sparse occupancy representation that bypasses the intermediate bird's eye view (BEV) projection and its explicit geometric priors
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出一种用于未来三维场景占用轨迹条件预测的新架构。与依赖变分自编码器（VAE）生成离散占用 token 的方法不同——这类方法本质上限制了表征能力——我们的方法以端到端方式直接从原始图像特征预测多帧未来占用。受 GPT、VGGT 等基础视觉与语言模型中基于注意力的 Transformer 架构取得成功的启发，我们采用稀疏占用表示，绕过中间鸟瞰图（BEV）投影及其显式几何先验。这一设计使 Transformer 能够更有效地捕捉时空依赖关系。通过同时避免离散 token 化的有限容量约束与 BEV 表示的结构性局限，我们的方法在 nuScenes 基准的 1–3 秒占用预测任务上取得最优性能，显著优于现有方法。此外，该方法展现出对场景动态的稳健理解能力，在任意未来轨迹条件下均能持续保持高精度。
英文原文摘要 1198 字符
This paper introduces a novel architecture for trajectory-conditioned forecasting of future 3D scene occupancy. In contrast to methods that rely on variational autoencoders (VAEs) to generate discrete occupancy tokens, which inherently limit representational capacity, our approach predicts multi-frame future occupancy in an end-to-end manner directly from raw image features. Inspired by the success of attention-based transformer architectures in foundational vision and language models such as GPT and VGGT, we employ a sparse occupancy representation that bypasses the intermediate bird's eye view (BEV) projection and its explicit geometric priors. This design allows the transformer to capture spatiotemporal dependencies more effectively. By avoiding both the finite-capacity constraint of discrete tokenization and the structural limitations of BEV representations, our method achieves state-of-the-art performance on the nuScenes benchmark for 1-3 second occupancy forecasting, outperforming existing approaches by a significant margin. Furthermore, it demonstrates robust scene dynamics understanding, consistently delivering high accuracy under arbitrary future trajectory conditioning.