跳到正文
arXiv cs.AI· Karthik Iyer·· 2 小时前AI 评分36

用训练中刻意保留的动作测试世界模型:零售场景中的保留计划、预测精度与定价规则

Testing World Models on Actions They Have Never Seen: Held-Out Plans, Forecast Accuracy and Price Rules in Retail

AI 导读

一项arXiv研究提出测试动作条件世界模型处理训练中未见动作的方法,在零售价格与促销计划场景中拟合了2,853个模型。结果显示预测精度仅在模型类别间有参考价值(秩相关0.32到0.66),同一模型版本间甚至会反转至-0.69;JEPA预训练等训练手段不降低新计划上的误差。

正文 · 原文

View PDF HTML (experimental)

Abstract:A world model is trusted to answer "what if": how its world responds to an action it has never seen. We give a test of that trust for any action-conditioned world model: actions held out of training by design and scored against a known answer; whether forecast accuracy predicts accuracy on those actions; and the same structure given to simpler models. We run it in retail, where the action is a price and promotion plan, across 2,853 fitted models. First, forecast accuracy is a fair guide between kinds of model (rank correlation with error on new plans 0.32 to 0.66) and a poor guide between versions of one world model: where merchants plan into high-demand weeks the link reverses, down to -0.69. Second, training the world model to learn the effect of actions (JEPA pretraining, a latent consistency check, adaptation at the plan date) leaves its error on new plans unchanged. Third, in generated worlds the recorded regular price and three price rules improve it. Given the same rules, a log-linear model is best when its curve can match the truth; in a world whose curve lies outside the rules, the world model with the rules is ahead of the log-linear model with the rules in 36 of 40 worlds, and does as well without its recurrent cell. On real data (Breakfast at the Frat, The Complete Journey and M5) the rules lower the world model's error on pure price cuts deeper than any it saw (0.63 to 0.56 on the Frat; 0.45 to 0.44, within chance, on The Complete Journey), but not on lasting price moves. Fourth, a small share of random test plans helps every model a little. In a generated dosing world, a stress test in a second domain, the first finding repeats, and the built-up drug level lowers the world model's error (2.86 to 1.91) though it still predicts only 0.12 to 0.32 of the true gain. The test carries to any domain where a world model is asked about actions outside its history.
Comments: 12 pages, 3 figures, 5 tables. Companion to arXiv:2611.20811
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.22537 [cs.LG]
  (or arXiv:2611.22537v1 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2611.22537

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Karthik Iyer [view email]
[v1] Tue, 17 Nov 2026 10:29:02 UTC (86 KB)

来源:arXiv cs.AI · arxiv.org