外观
论文arXiv 2609.211551 个月前
同一世界，不同知识：孤立审计何时误判世界模型修复
原文链接：Same World, Different Knowledge: When Isolated Audits Misjudge World-Model Repairs（arXiv 2609.21155 · 发布日期：2026-09-17）
Same World, Different Knowledge: When Isolated Audits Misjudge World-Model Repairs
论文提出信息接口审计来评估世界模型修复选择，并在仿真四旋翼模型预测控制中验证共享故障信息会改变修复优劣。
核心要点
- 孤立输入故障下偏好的修复，在部署模块共享故障信息时可能反而更差。
- 方法区分保真缺口与可用性缺口，用固定权重干预测量预测误差、输入依赖和闭环收益。
- 仿真四旋翼MPC中，质量/推力10%相反标定误差使成功率由69%降至8%，不确定性训练恢复至65%。
- 风重建可恢复控制收益但继承标定依赖；修复选择取决于部署中的信息路径。
验证范围与局限：验证范围为仿真四旋翼模型预测控制，涉及标定误差、风重建和扰动观测器。
英文摘要 (Abstract)
A repair favored under an isolated input fault can be inferior when deployed modules share the faulty information. We introduce an information-interface audit for world models, distinguishing fidelity gaps, where exact inputs become estimates, from availability gaps, where inputs are missing. Fixed-weight interventions measure prediction error, input dependence, and paired closed-loop benefit, including dependencies introduced by reconstruction. In simulated quadrotor model predictive control, coupled, opposite-sign 10% mass/thrust calibration errors reduce a physics-anchored model's success from 69% to 8%; uncertainty training restores 65%. Wind reconstruction recovers control benefit but inherits calibration dependence. For a positive calibration offset, reconstruction-only corruption favors uncertainty-trained reconstruction, whereas shared corruption favors the baseline. Acceleration diagnostics reveal compensation between reconstruction bias and nominal-model error, also observed with a disturbance observer. Repair selection therefore depends on the information paths used in deployment.
作者：Rui Min, Xianyao Li, Fang Xu, Sofiane Lachab, Jing Du
来源
本页事实对应的公开页面，以原文为准。