OpenTSLM TeeMoE：用于预测、上下文预测和推理的统一时间序列语言模型
OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and Reasoning
- Columbia University, USA（哥伦比亚大学）
- Stanford University, USA（斯坦福大学）
- Aionic Labs, Switzerland（Aionic Labs）
- Google Agentic Systems Lab, ETH Zürich, Switzerland（谷歌智能体系统实验室，苏黎世联邦理工学院）
- University of Pisa, Italy（比萨大学）
- National University of Singapore（新加坡国立大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出OpenTSLM TeeMoE，一种统一时间序列语言模型，通过LoRA混合专家控制器融合预测聚合、原生预测和时间分析三个低秩专家，在多个基准上取得前三性能。
AI中文摘要：
现实世界中的时间序列应用日益需要能够处理时间序列预测、上下文条件预测和基于语言的时间推理的模型。然而，当前的时间序列基础模型在这些能力上仍然碎片化：数值专家模型通常提供最强的预测，而基于语言的模型则提供更广泛的上下文理解和分析。一个核心挑战是在不降低各自性能的情况下统一这些异构能力。我们引入了OpenTSLM TeeMoE，一个通用型时间序列语言模型，它可以直接从观测到的时间序列进行预测，基于文本上下文和时间模式进行推理，并综合和优化来自外部数值预测专家的预测。我们在共享骨干网络上独立训练三个低秩专家，分别用于预测聚合、原生预测和时间分析。一个学习的LoRA混合专家控制器随后为每个请求加权其冻结的参数更新。我们提出的模型在广泛使用的时间序列预测、上下文条件预测和基于语言的时间推理基准上取得了强劲性能，在GIFT-Eval上按平均MASE排名位列前三，在Context is Key上按RCRPS位列前三，在TimeSeriesExam上按准确率位列前三。
英文摘要：
Real-world time-series applications increasingly require models that can handle time series forecasting, context-conditioned prediction, and language-based temporal reasoning. Yet current time-series foundation models remain fragmented across these capabilities: numerical specialists often provide the strongest forecasts, while language-based models offer broader contextual understanding and analysis. A central challenge is to unify these heterogeneous capabilities without reducing their individual performance. We introduce OpenTSLM TeeMoE, a generalist time-series language model that can forecast directly from observed time series, reason over textual context and temporal patterns, and synthesize and refine predictions from external numerical forecasting specialists. We independently train three low-rank experts for forecast aggregation, native forecasting, and temporal analysis over a shared backbone. A learned LoRA mixture-of-experts controller then weights their frozen parameter updates for each request. Our proposed model achieves strong performance on widely used benchmarks for time series forecasting, context-conditioned prediction, and language-based temporal reasoning, ranking among the top three on GIFT-Eval by mean MASE rank, Context is Key by RCRPS, and TimeSeriesExam by accuracy.
