Den-TP：面向轨迹预测的密度均衡数据筛选与评测框架
Den-TP: A Density-Balanced Data Curation and Evaluation Framework for Trajectory Prediction
把「场景密度」当作与数据集无关的交互复杂度代理，用梯度子模选择做密度均衡的数据子集筛选，并配套密度条件评测协议；做自动驾驶轨迹预测或任何长尾感知任务的人可以直接借用这套数据筛选与分桶评测思路，暴露被平均误差掩盖的高密度失效模式。
原文依据 applies a gradient-based submodular selection objective to choose representative samples within each region
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论自动驾驶中的轨迹预测传统上是从以模型为中心的视角来研究的。然而，现有数据集在场景密度上呈现强烈的长尾分布，常见的低密度场景占据主导，而安全关键的高密度场景则严重不足。这种不平衡限制了模型的鲁棒性，并且在标准评测对所有场景的误差取平均时掩盖了失效模式。我们从以数据为中心的视角重新审视轨迹预测，提出 Den-TP，一个面向密度感知的数据集筛选与评测框架。Den-TP 首先以智能体数量作为与数据集无关的交互复杂度代理，将数据划分为密度条件区域；随后应用基于梯度的子模选择目标，在每个区域内挑选代表性样本，同时显式地在不同密度之间重新平衡。所得子集将数据集规模缩减 50%，却保持了整体性能，并显著提升了高密度场景下的鲁棒性。我们进一步引入密度条件评测协议，揭示出被传统指标忽视的长尾失效模式。在 Argoverse 1 和 2 上使用最优模型进行的实验表明，鲁棒的轨迹预测不仅取决于数据规模，还取决于场景密度的均衡。
英文原文摘要 1338 字符
Trajectory prediction in autonomous driving has traditionally been studied from a model-centric perspective. However, existing datasets exhibit a strong long-tail distribution in scenario density, where common low-density cases dominate and safety-critical high-density cases are severely underrepresented. This imbalance limits model robustness and hides failure modes when standard evaluations average errors across all scenarios. We revisit trajectory prediction from a data-centric perspective and present Den-TP, a framework for density-aware dataset curation and evaluation. Den-TP first partitions data into density-conditioned regions using agent count as a dataset-agnostic proxy for interaction complexity. It then applies a gradient-based submodular selection objective to choose representative samples within each region while explicitly rebalancing across densities. The resulting subset reduces the dataset size by 50% yet preserves overall performance and significantly improves robustness in high-density scenarios. We further introduce density-conditioned evaluation protocols that reveal long-tail failure modes overlooked by conventional metrics. Experiments on Argoverse 1 and 2 with state-of-the-art models show that robust trajectory prediction depends not only on data scale, but also on balancing scenario density.