论文/arXiv 2610.09305·21 天前

Kuration SDK：通过数据策展解决 Virtual2Real 差距

论文针对动作条件世界模型的 Virtual2Real 差距，提出用数据策展与诊断属性度量，并在 CounterStrike 游戏数据上验证 Kuration SDK。

AI 整理

• 数据与采集
• 具身模型
• 评测与安全

原文链接：Kuration SDK: Addressing the Virtual2Real Gap via Data Curation（arXiv 2610.09305 · 发布日期：2026-10-07）

Kuration SDK: Addressing the Virtual2Real Gap via Data Curation

论文针对动作条件世界模型的 Virtual2Real 差距，提出用数据策展与诊断属性度量，并在 CounterStrike 游戏数据上验证 Kuration SDK。

核心要点

• 现有基准对通用动作条件世界模型训练信号有限，可玩性与 FVD、LPIPS、JEDi 等指标不一致。
• 作者提出在训练前策展原始游戏数据，并度量多种诊断属性，以更稳健地桥接差距。
• 在 CounterStrike 游戏数据上训练和评估扩散世界模型，确认定性可玩性与常用指标不对应。
• Kuration SDK 随论文开源，并在一个具体案例中帮助揭示相近指标下行为差异的根因。

验证范围与局限：验证范围限于 CounterStrike 游戏数据上的扩散世界模型及一个具体数据集案例。

英文摘要 (Abstract)

Benchmarks for measuring the quality of action-conditioned world models are still evolving and shifting away from visual similarity-based metrics to action-semantic and physically-grounded metrics. However, for domain and task-agnostic action-conditioned world model training, existing benchmarks provide a limited signal. By training and evaluating diffusion world models on CounterStrike gameplay data, we confirm that qualitative playability does not correspond with metrics such as FVD, LPIPS, and JEDi. We term this the Virtual2Real gap. We posit that, in lieu of reliable benchmarks, curating raw gameplay data and measuring a variety of diagnostic properties provides a more robust signal to bridge the gap, before the training even begins. We present several curation strategies and a general-purpose kit for physical AI data curation called Kuration SDK, which is being open-sourced with this paper. The SDK was instrumental in uncovering the root cause of the virtual2real gap in a specific case: why two world models trained on identical gameplay map, action and state distribution, behaved very differently when played in spite of having very similar LPIPS and FVD scores. Thus, Kuration SDK has the potential to uncover the root causes of Virtual2Real gap in specific datasets and accelerate development of sample-efficient training datasets.

作者：Nirmit V. Desai, Eric Song, M. Sengupta, Tejal Bedmutha, Siri Reddy, Sahiti Dharmavaram, Kunal Sawarkar

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.09305：Kuration SDK: Addressing the Virtual2Real Gap via Data Curation访问 2026-10-21