KairosVL: Orchestrating Time Series and Semantics for Unified Reasoning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914346800513024 |
|---|---|
| author | Si, Haotian Pei, Changhua He, Xiao Li, Zeyan Xie, Zhe Wang, Zexin Hu, Jiyao Yu, Zhaoyang Zhang, Tieying Pei, Dan Li, Jianhui Xie, Gaogang |
| author_facet | Si, Haotian Pei, Changhua He, Xiao Li, Zeyan Xie, Zhe Wang, Zexin Hu, Jiyao Yu, Zhaoyang Zhang, Tieying Pei, Dan Li, Jianhui Xie, Gaogang |
| contents | Driven by the increasingly complex and decision-oriented demands of time series analysis, we introduce the Semantic-Conditional Time Series Reasoning task, which extends conventional time series analysis beyond purely numerical modeling to incorporate contextual and semantic understanding. To further enhance the mode's reasoning capabilities on complex time series problems, we propose a two-round reinforcement learning framework: the first round strengthens the mode's perception of fundamental temporal primitives, while the second focuses on semantic-conditioned reasoning. The resulting model, KairosVL, achieves competitive performance across both synthetic and real-world tasks. Extensive experiments and ablation studies demonstrate that our framework not only boosts performance but also preserves intrinsic reasoning ability and significantly improves generalization to unseen scenarios. To summarize, our work highlights the potential of combining semantic reasoning with temporal modeling and provides a practical framework for real-world time series intelligence, which is in urgent demand. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_20494 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | KairosVL: Orchestrating Time Series and Semantics for Unified Reasoning Si, Haotian Pei, Changhua He, Xiao Li, Zeyan Xie, Zhe Wang, Zexin Hu, Jiyao Yu, Zhaoyang Zhang, Tieying Pei, Dan Li, Jianhui Xie, Gaogang Artificial Intelligence Driven by the increasingly complex and decision-oriented demands of time series analysis, we introduce the Semantic-Conditional Time Series Reasoning task, which extends conventional time series analysis beyond purely numerical modeling to incorporate contextual and semantic understanding. To further enhance the mode's reasoning capabilities on complex time series problems, we propose a two-round reinforcement learning framework: the first round strengthens the mode's perception of fundamental temporal primitives, while the second focuses on semantic-conditioned reasoning. The resulting model, KairosVL, achieves competitive performance across both synthetic and real-world tasks. Extensive experiments and ablation studies demonstrate that our framework not only boosts performance but also preserves intrinsic reasoning ability and significantly improves generalization to unseen scenarios. To summarize, our work highlights the potential of combining semantic reasoning with temporal modeling and provides a practical framework for real-world time series intelligence, which is in urgent demand. |
| title | KairosVL: Orchestrating Time Series and Semantics for Unified Reasoning |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2602.20494 |