TimesJev: Thinking Fast on Time Series with System 1 Sufficiency
作者: Chenghao Liu, Wei Fan, Jindong Han, Dongjie Wang, Aniket Mahanti, Pengyang Wang, Hao Liu
分类: cs.AI
发布日期: 2026-10-19
💡 一句话要点
提出TimesJev以解决时间序列推理中的延迟问题
🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 时间序列推理 System 1模型 压缩表示 推理效率 机器学习
📋 核心要点
- 现有时间序列推理模型通常需要长时间的推理链，导致推理延迟和资源消耗高。
- 本文提出TimesJev，通过压缩时间序列表示并保留关键频率模式，提升推理效率。
- 实验表明，TimesJev在推理性能上提升超过62%，且推理时间减少62.9%。
📝 摘要（中文）
大型语言模型（LLMs）在时间序列推理中展现出强大的潜力，但现有模型通常需要长时间的推理链，导致推理延迟显著。为此，本文首次提出了一种时间序列的System 1模型——TimesJev。我们首先形式化了时间序列推理中的System 1充分性，进而提出了一种压缩表示方法，保留关键频率模式。通过结合充分性散度和可实现性散度的目标，我们优化了时间序列的压缩过程，并通过门控机制学习选择合适的表示。实验结果表明，TimesJev在13个数据集上表现优于现有推理模型，推理时间平均减少62.9%。
🔬 方法详解
问题定义：本文旨在解决时间序列推理中的延迟问题，现有模型在处理长时间序列时需要消耗大量的计算资源和时间，影响实际应用。
核心思路：我们提出了一种新的时间序列System 1模型——TimesJev，通过压缩时间序列的表示来提高推理速度，同时保留关键的频率模式，以适应System 1模型的需求。
技术框架：整体架构包括两个主要阶段：首先是时间序列的压缩表示学习，其次是将压缩后的表示与System 1模型的嵌入空间对齐。
关键创新：最重要的创新在于引入了充分性散度和可实现性散度的结合，形成了一种新的压缩目标，使得压缩后的表示能够有效支持快速推理。
关键设计：在模型设计中，采用了门控机制来选择合适的时间序列表示，并通过优化损失函数来平衡信息保留与模型兼容性。
🖼️ 关键图片
📊 实验亮点
在13个数据集上的实验结果显示，TimesJev在推理性能上超越了现有的推理模型，平均提升超过62%，同时推理时间减少62.9%，显著提高了效率。
🎯 应用场景
TimesJev的研究成果在金融预测、气象数据分析和智能监控等领域具有广泛的应用潜力。通过提高时间序列推理的效率，能够更快速地做出决策，提升系统的响应能力和准确性，具有重要的实际价值和未来影响。
📄 摘要（原文）
Large language models (LLMs) have recently shown strong potential for time series reasoning, while models that reason over time series are usually with long chains of thought. However, this process often consumes thousands of tokens, directly leading to substantial latency during inference. In contrast, System 1 models such as Jev make decisions instantly and can naturally be an alternative, but raw time series are not compatible with the embedding space of these models, thus hindering the reasoning performance, especially when input series contain thousands of numeric points. In this paper, we make a very early attempt to propose a time series System 1 model, namely TimesJev. Specifically, we first formalize System 1 sufficiency in time series reasoning, which motivates a compressed representation for time series with key frequency patterns preserved. To construct this representation, we derive a dimension reduction objective that combines (i) a sufficiency divergence that measures the core information lost during compression, and (ii) a realizability divergence that quantifies the incompatibility of the System 1 model to the optimal prediction given this representation. We optimize this objective by learning time series compression with a gating mechanism for sufficiency, and then aligning the selected representation with the embedding space for realizability. Finally, experiments on 13 datasets demonstrate that TimesJev outperforms reasoning models by over 62% with 62.9% less inference time than their fine-tuned versions on average.