Welcome to the TS_Instruct_QA_Gold dataset
This dataset is intended to evaluate time-series reasoning.
The dataset consists of real-world time-series with synthetic text.
The dataset was human evaluated for correctness
@misc{quinlan2025chattsenhancingmultimodalreasoning,
title={Chat-TS: Enhancing Multi-Modal Reasoning Over Time-Series and Natural Language Data},
author={Paul Quinlan and Qingguo Li and Xiaodan Zhu},
year={2025},
eprint={2503.10883},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2503.10883},
}