发表机构
University of California San Diego; University of Southern California; Aether AI（加州大学圣迭戈分校; 南加州大学; Aether AI）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
TimeBraid通过交错残差注意力层统一预训练语言与时间序列模型，利用共享表示和联合训练，在多种基准上实现与更大模型相当的竞争力。
AI 中文摘要
我们提出了TimeBraid，一系列统一的时间序列与语言模型，通过交错全局残差注意力层，将预训练语言模型与预训练时间序列基础模型对齐。每个模型一方面继承了知识、指令遵循和推理能力，另一方面继承了连续信号感知和零样本预测能力，并在共享表示空间中将两者融合，使两种模态都能被理解和生成。我们研究了使这种统一建模有效的设计选择：在哪里对齐两个表示空间，如何将语言锚定在时间结构中，如何平衡理解与生成，以及如何保持联合优化的稳定性。由此产生的方案结合了针对多样时间序列和文本任务的统一提示方案、稳定的联合训练，以及来自220万条精选序列-文本对和490万条指令微调样本的监督。在涵盖时间序列感知、理解、推理以及上下文辅助和单模态预测的基准测试中，TimeBraid与规模大得多的通用模型和任务特定模型保持竞争力。
英文摘要
We present TimeBraid, a series of unified time-series and language models that align pretrained language models and pretrained time-series foundation models through interleaved global residual attention layers. Each model inherits knowledge, instruction following, and reasoning from one side, continuous-signal perception and zero-shot forecasting from the other, and fuses the two in a shared representation space where both modalities are understood and generated. We study the design choices that make such unified modeling work: where to align the two representation spaces, how to ground language in temporal structure, how to balance understanding with generation, and how to keep joint optimization stable. The resulting recipe combines a unified prompting scheme for diverse time-series and text tasks, stabilized joint training, and supervision from 2.2M curated series--text pairs and 4.9M instruction-tuning samples. Across benchmarks spanning time-series perception, understanding, reasoning, and both context-aided and unimodal forecasting, TimeBraid remains competitive with far larger general-purpose models and task-specific counterparts.
Comments57 pages
