LLaTiSA：從視覺感知到語義的難度分層時間序列推理

LLaTiSA: Towards Difficulty-Stratified Time Series Reasoning from Visual Perception to Semantics

April 19, 2026

作者: Yueyang Ding, HaoPeng Zhang, Rui Dai, Yi Wang, Tianyu Zong, Kaikui Liu, Xiangxiang Chu

cs.AI

摘要

對時間序列的全面理解仍是大型語言模型面臨的重大挑戰。當前研究受限於碎片化的任務定義與存在固有模糊性的基準測試，阻礙了嚴謹評估及統一時間序列推理模型的發展。為彌合此差距，我們通過四級認知複雜度遞增的分類法，正式定義了時間序列推理。我們推出分層時間序列推理數據集HiTSR，包含8.3萬個樣本，涵蓋多樣化任務組合並附有經驗證的思維鏈軌跡。基於HiTSR，我們提出LLaTiSA——一個強大的時間序列推理模型，通過將可視化模式與精度校準的數值表格相融合，增強視覺語言模型的時序感知能力。採用多階段課程精調策略後，LLaTiSA在各類時間序列推理任務及真實場景中展現出卓越性能與穩健的分布外泛化能力。代碼已開源於：https://github.com/RainingNovember/LLaTiSA。

English

Comprehensive understanding of time series remains a significant challenge for Large Language Models (LLMs). Current research is hindered by fragmented task definitions and benchmarks with inherent ambiguities, precluding rigorous evaluation and the development of unified Time Series Reasoning Models(TSRMs). To bridge this gap, we formalize Time Series Reasoning (TSR) via a four-level taxonomy of increasing cognitive complexity. We introduce HiTSR, a hierarchical time series reasoning dataset comprising 83k samples with diverse task combinations and verified Chain-of-Thought (CoT) trajectories. Leveraging HiTSR, we propose LLaTiSA, a strong TSRM that integrates visualized patterns with precision-calibrated numerical tables to enhance the temporal perception of Vision-Language Models (VLMs). Through a multi-stage curriculum fine-tuning strategy, LLaTiSA achieves superior performance and exhibits robust out-of-distribution generalization across diverse TSR tasks and real-world scenarios. Our code is available at https://github.com/RainingNovember/LLaTiSA.