发表机构
Korea Institute of Industrial Technology; Hanyang University（韩国生产技术研究院; 汉阳大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究发现时间序列基础模型在反事实输入下对动态系统的识别存在衰减，提出合成强迫系统微调修复方法，恢复响应幅度并优于结构无关识别。
AI 中文摘要
协变量感知的时间序列基础模型（TSFMs）承诺为装备了仪器的工厂提供无需训练的假设分析答案：即不同未来输入将导致的输出变化。我们在具有精确反事实的强迫工程系统上对此进行了测试，将Chronos-2、TimesFM-2.5和TabPFN-TS与拟合于相同上下文的经典系统识别方法进行了比较。通过其默认的协变量接口，TimesFM-2.5和TabPFN-TS是无记忆的：输入变化对输出的预测影响是该变化的同时间函数（TimesFM-2.5的R²=1.000）。Chronos-2能在上下文中识别动态特性，但会对其衰减。其预测效果仅为真实效果的0.33-0.80，恢复的脉冲响应形状不正确，且在单自由度振荡器上的误差在8192个上下文样本时趋于0.57，而拟合256个样本的ARX误差达到0.02。推理时的上下文抖动可在无需训练的情况下降低所有六个合成类别的假设分析误差。在合成强迫系统上进行26分钟的微调恢复了响应幅度（灵敏度0.83-0.96），并在Wiener-Hammerstein和保留的摩擦类别上优于结构无关的识别方法。在相同数据上训练的专业上下文识别器接近该性能，因此强迫系统数据带来了大部分增益。在四个实测工厂中的三个上，经典识别方法仍然明显更好，且微调模型失去了部分单变量预测技能。配对的反事实输入，连同实测记录上打乱的未来输入，测试了两个属性：协变量接口是否能表示动态特性，以及预训练先验是否覆盖工厂的时间尺度。只有反事实对暴露了衰减问题。
英文摘要
Covariate-aware time-series foundation models (TSFMs) promise training-free what-if answers for instrumented plants: the change in output that a different future input would cause. We test this on forced engineering systems with exact counterfactuals, comparing Chronos-2, TimesFM-2.5 and TabPFN-TS with classical system identification fitted to the same context. Through their default covariate interfaces, TimesFM-2.5 and TabPFN-TS are memoryless: the predicted effect of an input change is a same-time function of that change ($R^2 = 1.000$ for TimesFM-2.5). Chronos-2 identifies dynamics in context but attenuates them. Its predicted effect is 0.33-0.80 of the true effect, its recovered impulse response has the wrong shape, and its error on a one-degree-of-freedom oscillator levels off at 0.57 with 8192 context samples, where ARX fitted to 256 samples reaches 0.02. Context dither at inference lowers the what-if error on all six synthetic classes without training. A 26-minute fine-tune on synthetic forced systems restores the response magnitude (sensitivity 0.83-0.96) and outperforms structure-agnostic identification on Wiener-Hammerstein and a held-out friction class. A specialised in-context identifier trained on the same data comes close, so the forced-system data carry most of the gain. On three of four measured plants classical identification remains clearly better, and the fine-tuned model loses part of its univariate forecasting skill. Paired counterfactual inputs, together with shuffled future inputs on measured records, test two properties: whether the covariate interface can represent dynamics and whether the pretraining prior covers the plant's time scale. Only the counterfactual pairs expose the attenuation.
Comments12 pages, 4 figures, 5 tables
