Chart-to-Text: A Large-Scale Benchmark for Chart Summarization

Kantharaj, Shankar; Leong, Rixie Tiffany; Lin, Xin; Masry, Ahmed; Thakkar, Megh; Hoque, Enamul; Joty, Shafiq

doi:10.18653/v1/2022.acl-long.277

Cited by 36 publications

(42 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, participants completed the demographic section. Here, they reported their age range (e.g., [18][19][20][21][22][23][24], their current education level (e.g., "Less than high school"), and their experience with charts and reading. This experience was captured through both the frequency (e.g., "Every day") as well as the context in which they engaged with the material (e.g., Government reports).…”

Section: Methodsmentioning

confidence: 99%

“…Besides serving as input or output for visual analysis, systems have begun to incorporate natural language-generated text along with their visualization responses to help describe key insights to the user. For example, a variety of research systems [6,21,40,43,46] and tools such as Tableau's Summary Card [51], and Power BI [38] produce natural language captions that summarize statistics and trends depicted by the chart.…”

Section: Visualization + Text For Analysis and Storytellingmentioning

confidence: 99%

See 1 more Smart Citation

Striking a Balance: Reader Takeaways and Preferences when Integrating Text and Charts

Stokes,

Setlur,

Cogley

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Visualization + Text For Analysis and Storytellingmentioning

confidence: 99%

Striking a Balance: Reader Takeaways and Preferences when Integrating Text and Charts

Stokes,

Setlur,

Cogley

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…To avoid too simple and repetitive structure, we analyzed three different corpora of summary sentences for charts and curated several templates for each message type with varying syntactic structures. To find various variations of templates, we analyzed the dataset shared by Kanthara et al in their extended work on Chart-to-Text [45]. We randomly pick a different template every time a particular category of message appears to enhance naturalness and diversity in lexicons.…”

Section: Template-based (Moderate Length)mentioning

confidence: 99%

SeeChart: Enabling Accessible Visualizations Through Interactive Natural Language Interface For People with Visual Impairments

Alam

Islam

Hoque

2023

Proceedings of the 28th International Conference on Intelligent User Interfaces

View full text Add to dashboard Cite

“…Ultimately, two model variants are developed: ChartVLM-Base-7.3B (0.3B chart image encoder & base decoder + 7B auxiliary decoder) and ChartVLM-Large-14.3B (1.3B chart image encoder & base decoder + 13B auxiliary decoder). All the data we used during fine-tuning stage comes from ChartQA [23], PlotQA [25], Chart2Text [14], and SimChart9K [34]. Besides, the ChartVLM is trained using 32 × NVIDIA Tesla A100.…”

Section: Cascaded Decoders Designmentioning

confidence: 99%

Progress and Perspective of Artificial Intelligence and Machine Learning of Prediction in Anesthesiology

Xia

Jiang

2021

J. Shanghai Jiaotong Univ. (Sci.)

View full text Add to dashboard Cite

Charts are common in literature across different scientific fields, conveying rich information easily accessible to readers. Current chart-related tasks focus on either chart perception which refers to extracting information from the visual charts, or performing reasoning given the extracted data, e.g. in a tabular form. In this paper, we aim to establish a unified and label-efficient learning paradigm for joint perception and reasoning tasks, which can be generally applicable to different downstream tasks, beyond the question-answering task as specifically studied in peer works. Specifically, StructChart first reformulates the chart information from the popular tubular form (specifically linearized CSV) to the proposed Structured Triplet Representations (STR), which is more friendly for reducing the task gap between chart perception and reasoning due to the employed structured information extraction for charts. We then propose a Structuring Chart-oriented Representation Metric (SCRM) to quantitatively evaluate the performance for the chart perception task. To enrich the dataset for training, we further explore the possibility of leveraging the Large Language Model (LLM), enhancing the chart diversity in terms of both chart visual style and its statistical information. Extensive experiments are conducted on various chart-related tasks, demonstrating the effectiveness and promising potential for a unified chart perception-reasoning paradigm to push the frontier of chart understanding. Our simulation chart dataset is available at: https://github.com/UniModal4Reasoning/SimChart9K, and the pre-training code will be released for reproduction.

show abstract

Chart-to-Text: A Large-Scale Benchmark for Chart Summarization

Cited by 36 publications

References 0 publications

Striking a Balance: Reader Takeaways and Preferences when Integrating Text and Charts

Striking a Balance: Reader Takeaways and Preferences when Integrating Text and Charts

SeeChart: Enabling Accessible Visualizations Through Interactive Natural Language Interface For People with Visual Impairments

Progress and Perspective of Artificial Intelligence and Machine Learning of Prediction in Anesthesiology

Contact Info

Product

Resources

About