Chart Question Answering from Real-World Analytical Narratives
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908430364573696 |
|---|---|
| author | Hutchinson, Maeve Jianu, Radu Slingsby, Aidan Wood, Jo Madhyastha, Pranava |
| author_facet | Hutchinson, Maeve Jianu, Radu Slingsby, Aidan Wood, Jo Madhyastha, Pranava |
| contents | We present a new dataset for chart question answering (CQA) constructed from visualization notebooks. The dataset features real-world, multi-view charts paired with natural language questions grounded in analytical narratives. Unlike prior benchmarks, our data reflects ecologically valid reasoning workflows. Benchmarking state-of-the-art multimodal large language models reveals a significant performance gap, with GPT-4.1 achieving an accuracy of 69.3%, underscoring the challenges posed by this more authentic CQA setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_01627 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Chart Question Answering from Real-World Analytical Narratives Hutchinson, Maeve Jianu, Radu Slingsby, Aidan Wood, Jo Madhyastha, Pranava Computation and Language We present a new dataset for chart question answering (CQA) constructed from visualization notebooks. The dataset features real-world, multi-view charts paired with natural language questions grounded in analytical narratives. Unlike prior benchmarks, our data reflects ecologically valid reasoning workflows. Benchmarking state-of-the-art multimodal large language models reveals a significant performance gap, with GPT-4.1 achieving an accuracy of 69.3%, underscoring the challenges posed by this more authentic CQA setting. |
| title | Chart Question Answering from Real-World Analytical Narratives |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2507.01627 |