arXiv cs.AI· Chenghao Liu, Wei Fan, Jindong Han, Dongjie Wang, Aniket Mahanti, Pengyang Wang, Hao Liu·· 4 小时前AI 评分40
TimesJev:用 System 1 充分性实现时间序列快速推理
TimesJev: Thinking Fast on Time Series with System 1 Sufficiency
AI 导读
arXiv 论文提出时间序列 System 1 模型 TimesJev,通过保留关键频率模式的压缩表示,将时间序列对齐到 System 1 模型的嵌入空间,避免长链思考带来的数千 token 推理延迟。该方法结合充分性散度与可实现性散度构建降维目标,用门控机制学习压缩。在 13 个数据集上,TimesJev 比推理模型高逾 62%,推理时间比其微调版本平均减少 62.9%。
正文 · 原文
Abstract: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.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.26002 [cs.AI] |
| (or arXiv:2610.26002v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.26002 arXiv-issued DOI via DataCite |
Submission history
From: Chenghao Liu [view email]
[v1]
Mon, 19 Oct 2026 08:26:07 UTC (364 KB)
[v2]
Sat, 31 Oct 2026 13:14:09 UTC (365 KB)
[v3]
Fri, 13 Nov 2026 12:29:10 UTC (365 KB)
来源:arXiv cs.AI · arxiv.org