用注入知识的方法，让大模型更懂时间序列诊断。
SenTSR-Bench: Thinking with Injected Knowledge for Time-Series Reasoning
- 将时序模型的洞察注入通用大模型推理过程
- 在多个数据集上超越现有方法7.9%至26.1%
- 无需人工标注，靠强化学习自动获取高质量推理轨迹
时间序列诊断推理在诸多应用中至关重要，但现有方法存在明显短板：通用大语言模型（GRLMs）具备强推理能力却缺乏领域知识，而微调过的时间序列大模型（TSLMs）虽能理解复杂模式，却难以泛化应对复杂问题。为此，我们提出一种混合知识注入框架，将TSLM生成的见解直接嵌入GRLM的推理链中，实现兼具领域知识与泛化能力的时间序列推理。由于知识注入训练数据收集成本高，我们进一步采用基于可验证奖励的强化学习（RLVR）方法，无需人工监督即可生成富含知识的推理轨迹，并高效迁移至GRLM完成知识注入。同时，我们发布了SenTSR-Bench，一个基于真实工业运行数据的多变量时间序列诊断推理基准。在SenTSR-Bench及其他公开数据集上，本方法相较TSLMs提升9.1%-26.1%，相较GRLMs提升7.9%-22.4%，提供稳健且情境感知的诊断洞察。
原文摘要 · Abstract (English)
Time-series diagnostic reasoning is essential for many applications, yet existing solutions face a persistent gap: general reasoning large language models (GRLMs) possess strong reasoning skills but lack the domain-specific knowledge to understand complex time-series patterns. Conversely, fine-tuned time-series LLMs (TSLMs) understand these patterns but lack the capacity to generalize reasoning for more complicated questions. To bridge this gap, we propose a hybrid knowledge-injection framework that injects TSLM-generated insights directly into GRLM's reasoning trace, thereby achieving strong time-series reasoning with in-domain knowledge. As collecting data for knowledge injection fine-tuning is costly, we further leverage a reinforcement learning-based approach with verifiable rewards (RLVR) to elicit knowledge-rich traces without human supervision, then transfer such an in-domain thinking trace into GRLM for efficient knowledge injection. We further release SenTSR-Bench, a multivariate time-series-based diagnostic reasoning benchmark collected from real-world industrial operations. Across SenTSR-Bench and other public datasets, our method consistently surpasses TSLMs by 9.1%-26.1% and GRLMs by 7.9%-22.4%, delivering robust, context-aware time-series diagnostic insights.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。