English

DataTales: A Benchmark for Real-World Intelligent Data Narration

Artificial Intelligence 2025-08-26 v2

Abstract

We introduce DataTales, a novel benchmark designed to assess the proficiency of language models in data narration, a task crucial for transforming complex tabular data into accessible narratives. Existing benchmarks often fall short in capturing the requisite analytical complexity for practical applications. DataTales addresses this gap by offering 4.9k financial reports paired with corresponding market data, showcasing the demand for models to create clear narratives and analyze large datasets while understanding specialized terminology in the field. Our findings highlights the significant challenge that language models face in achieving the necessary precision and analytical depth for proficient data narration, suggesting promising avenues for future model development and evaluation methodologies.

Keywords

Cite

@article{arxiv.2410.17859,
  title  = {DataTales: A Benchmark for Real-World Intelligent Data Narration},
  author = {Yajing Yang and Qian Liu and Min-Yen Kan},
  journal= {arXiv preprint arXiv:2410.17859},
  year   = {2025}
}

Comments

Accepted at EMNLP 2024 (main conference, long paper)

R2 v1 2026-06-28T19:32:52.575Z