MMTS-BENCH: A Comprehensive Benchmark for Time Series Understanding and Reasoning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918328726978560 |
|---|---|
| author | Yin, Yao Xiao, Zhenyu Li, Musheng Liu, Yiwen Nan, Sutong He, Yiting Wang, Ruiqi Zhang, Zhenwei Liao, Qingmin Gu, Yuantao |
| author_facet | Yin, Yao Xiao, Zhenyu Li, Musheng Liu, Yiwen Nan, Sutong He, Yiting Wang, Ruiqi Zhang, Zhenwei Liao, Qingmin Gu, Yuantao |
| contents | Time series data are central to domains such as finance, healthcare, and cloud computing, yet existing benchmarks for evaluating various large language models (LLMs) on temporal tasks remain scattered and unsystematic. To bridge this gap, we introduce MMTS-BENCH, a comprehensive multimodal benchmark built upon a hierarchical taxonomy of time-series tasks, spanning structural awareness, feature analysis, temporal reasoning, sequence matching and cross-modal alignment. MMTS-BENCH comprises 2,424 time series question answering (TSQA) pairs across 4 subsets: Base, InWild, Match, and Align, generated through a progressive real-world QA framework and modular synthetic data construction. We conduct extensive evaluations on closed-source, open-source LLMs and existing time series adapted large language models (TS-LLMs), revealing that: (1) TS-LLMs significantly lag behind general-purpose LLMs in cross-domain generalization, (2) LLMs show weaknesses in local tasks compared to global tasks, (3) chain-of-thought (CoT) reasoning and multimodal integration substantially improve performance, and (4) the dominant factor in existing TS-LLMs remains the backbone network capability rather than the time series encoder design. MMTS-BENCH not only provides a rigorous evaluation framework but also offers clear directions for advancing LLMs toward robust, interpretable, and generalizable time-series reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_08588 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MMTS-BENCH: A Comprehensive Benchmark for Time Series Understanding and Reasoning Yin, Yao Xiao, Zhenyu Li, Musheng Liu, Yiwen Nan, Sutong He, Yiting Wang, Ruiqi Zhang, Zhenwei Liao, Qingmin Gu, Yuantao Databases Time series data are central to domains such as finance, healthcare, and cloud computing, yet existing benchmarks for evaluating various large language models (LLMs) on temporal tasks remain scattered and unsystematic. To bridge this gap, we introduce MMTS-BENCH, a comprehensive multimodal benchmark built upon a hierarchical taxonomy of time-series tasks, spanning structural awareness, feature analysis, temporal reasoning, sequence matching and cross-modal alignment. MMTS-BENCH comprises 2,424 time series question answering (TSQA) pairs across 4 subsets: Base, InWild, Match, and Align, generated through a progressive real-world QA framework and modular synthetic data construction. We conduct extensive evaluations on closed-source, open-source LLMs and existing time series adapted large language models (TS-LLMs), revealing that: (1) TS-LLMs significantly lag behind general-purpose LLMs in cross-domain generalization, (2) LLMs show weaknesses in local tasks compared to global tasks, (3) chain-of-thought (CoT) reasoning and multimodal integration substantially improve performance, and (4) the dominant factor in existing TS-LLMs remains the backbone network capability rather than the time series encoder design. MMTS-BENCH not only provides a rigorous evaluation framework but also offers clear directions for advancing LLMs toward robust, interpretable, and generalizable time-series reasoning. |
| title | MMTS-BENCH: A Comprehensive Benchmark for Time Series Understanding and Reasoning |
| topic | Databases |
| url | https://arxiv.org/abs/2602.08588 |