Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2510.02410 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912905479323648 |
|---|---|
| author | Langer, Patrick Kaar, Thomas Rosenblattl, Max Xu, Maxwell A. Chow, Winnie Maritsch, Martin Jakob, Robert Wang, Ning Liu, Juncheng Verma, Aradhana Han, Brian Kim, Daniel Seung Chubb, Henry Ceresnak, Scott Zahedivash, Aydin Sandhu, Alexander Tarlochan Singh Rodriguez, Fatima McDuff, Daniel Fleisch, Elgar Aalami, Oliver Barata, Filipe Schmiedmayer, Paul |
| author_facet | Langer, Patrick Kaar, Thomas Rosenblattl, Max Xu, Maxwell A. Chow, Winnie Maritsch, Martin Jakob, Robert Wang, Ning Liu, Juncheng Verma, Aradhana Han, Brian Kim, Daniel Seung Chubb, Henry Ceresnak, Scott Zahedivash, Aydin Sandhu, Alexander Tarlochan Singh Rodriguez, Fatima McDuff, Daniel Fleisch, Elgar Aalami, Oliver Barata, Filipe Schmiedmayer, Paul |
| contents | LLMs have emerged as powerful tools for interpreting multimodal data. In medicine, they hold particular promise for synthesizing large volumes of clinical information into actionable insights and digital health applications. Yet, a major limitation remains their inability to handle time series. To overcome this gap, we present OpenTSLM, a family of Time Series Language Models (TSLMs) created by integrating time series as a native modality to pretrained LLMs, enabling reasoning over multiple time series of any length. We investigate two architectures for OpenTSLM. The first, OpenTSLM-SoftPrompt, models time series implicitly by concatenating learnable time series tokens with text tokens via soft prompting. Although parameter-efficient, we hypothesize that explicit time series modeling scales better and outperforms implicit approaches. We thus introduce OpenTSLM-Flamingo, which integrates time series with text via cross-attention. We benchmark both variants against baselines that treat time series as text tokens or plots, across a suite of text-time-series Chain-of-Thought (CoT) reasoning tasks. We introduce three datasets: HAR-CoT, Sleep-CoT, and ECG-QA-CoT. Across all, OpenTSLM models outperform baselines, reaching 69.9 F1 in sleep staging and 65.4 in HAR, compared to 9.05 and 52.2 for finetuned text-only models. Notably, even 1B-parameter OpenTSLM models surpass GPT-4o (15.47 and 2.95). OpenTSLM-Flamingo matches OpenTSLM-SoftPrompt in performance and outperforms on longer sequences, while maintaining stable memory requirements. By contrast, SoftPrompt grows exponentially in memory with sequence length, requiring around 110 GB compared to 40 GB VRAM when training on ECG-QA with LLaMA-3B. Expert reviews by clinicians find strong reasoning capabilities exhibited by OpenTSLMs on ECG-QA. To facilitate further research, we provide all code, datasets, and models open-source. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_02410 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | OpenTSLM: Time-Series Language Models for Reasoning over Multivariate Medical Text- and Time-Series Data Langer, Patrick Kaar, Thomas Rosenblattl, Max Xu, Maxwell A. Chow, Winnie Maritsch, Martin Jakob, Robert Wang, Ning Liu, Juncheng Verma, Aradhana Han, Brian Kim, Daniel Seung Chubb, Henry Ceresnak, Scott Zahedivash, Aydin Sandhu, Alexander Tarlochan Singh Rodriguez, Fatima McDuff, Daniel Fleisch, Elgar Aalami, Oliver Barata, Filipe Schmiedmayer, Paul Machine Learning LLMs have emerged as powerful tools for interpreting multimodal data. In medicine, they hold particular promise for synthesizing large volumes of clinical information into actionable insights and digital health applications. Yet, a major limitation remains their inability to handle time series. To overcome this gap, we present OpenTSLM, a family of Time Series Language Models (TSLMs) created by integrating time series as a native modality to pretrained LLMs, enabling reasoning over multiple time series of any length. We investigate two architectures for OpenTSLM. The first, OpenTSLM-SoftPrompt, models time series implicitly by concatenating learnable time series tokens with text tokens via soft prompting. Although parameter-efficient, we hypothesize that explicit time series modeling scales better and outperforms implicit approaches. We thus introduce OpenTSLM-Flamingo, which integrates time series with text via cross-attention. We benchmark both variants against baselines that treat time series as text tokens or plots, across a suite of text-time-series Chain-of-Thought (CoT) reasoning tasks. We introduce three datasets: HAR-CoT, Sleep-CoT, and ECG-QA-CoT. Across all, OpenTSLM models outperform baselines, reaching 69.9 F1 in sleep staging and 65.4 in HAR, compared to 9.05 and 52.2 for finetuned text-only models. Notably, even 1B-parameter OpenTSLM models surpass GPT-4o (15.47 and 2.95). OpenTSLM-Flamingo matches OpenTSLM-SoftPrompt in performance and outperforms on longer sequences, while maintaining stable memory requirements. By contrast, SoftPrompt grows exponentially in memory with sequence length, requiring around 110 GB compared to 40 GB VRAM when training on ECG-QA with LLaMA-3B. Expert reviews by clinicians find strong reasoning capabilities exhibited by OpenTSLMs on ECG-QA. To facilitate further research, we provide all code, datasets, and models open-source. |
| title | OpenTSLM: Time-Series Language Models for Reasoning over Multivariate Medical Text- and Time-Series Data |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2510.02410 |