iiinova — where your digital self keeps evolving

Interactive Data Lab

@idl.uw.edu

Visualization & data analysis research at the University of Washington. In a prior life was the Stanford Vis Group. idl.uw.edu

Posts Replies Media

Reposted by Interactive Data Lab

Jeffrey Heer @jheer.org · 29/04/2025

The Mosaic architecture for database-backed interactive visualization just hit 1,000 stars on GitHub! Thank you to all who have participated and contributed!

Screenshot from GitHub showing 1k stars for the Mosaic project, an extensible framework for linking database and interactive views.

0312

Interactive Data Lab @idl.uw.edu · 26/11/2024

DracoGPT was led by Will Wang, in collaboration with Mitchell Gordon, Leilani Battle, and @jheer.org. For more, see the VIS'24 paper "DracoGPT: Extracting Visualization Design Preferences from Large Language Models" idl.uw.edu/papers/draco...

010

Interactive Data Lab @idl.uw.edu · 26/11/2024

For value comparison tasks, GPT-4 Turbo mostly aligns with human performance data. But for summary tasks, the GPT responses are uncorrelated and likely unhelpful! This may be due to much research and punditry on value comparison, but less attention to aggregate perception, biasing LLM training data.

110

Interactive Data Lab @idl.uw.edu · 26/11/2024

We applied DracoGPT using experimental data from Younghoon Kim et al. (EuroVis'18) that spans a variety of plots, data distributions, and tasks (comparing individual values vs. aggregate properties). The plots show chart "scores" from two GPT-4 Turbo prompting methods vs. human performance data.

Chart costs conditioned on task type, comparing Kim et al.’s experimental results to DracoGPT-Rank (A) and DracoGPT-Recommend Vega-Lite (B). For value tasks, results from both GPT4-Turbo models moderately correlate with a model fit to human performance data, but do not significantly correlate for summary tasks.

100

Interactive Data Lab @idl.uw.edu · 26/11/2024

DracoGPT prompts an LLM to rank or recommend charts, and uses the results to create new training pairs. It then learns constraint weights to model the design preferences expressed by the LLM. We can compare different fitted knowledge bases to see how their design choices align or diverge.

100

Interactive Data Lab @idl.uw.edu · 26/11/2024

Draco represents visualizations as a set of facts (about data & encodings) and constraints (preferences about encodings, scales, etc - such as if a bar chart has a zero baseline). From labeled chart pairs, we learn constraint weights ("costs") that balance competing constraints to "score" a chart.

100

Interactive Data Lab @idl.uw.edu · 26/11/2024

People are now using LLMs to create charts and graphs. How might we assess the quality and consistency of the results? DracoGPT is a method that fits a visualization knowledge base (Draco) to LLM responses, enabling comparison across models, prompts, and results from human subjects experiments.

Overview of the DracoGPT-Rank pipeline. (1) User provides prompt templates for an LLM to rank chart pairs; (2) Draco featurizes charts and produces feature vectors consisting of constraint counts; (3) Draco learns constraint weights over LLM-labeled chart pairs by fitting a RankSVM model; (4) The fitted Draco model can be applied to score charts. Results at each stage of the pipeline afford insight into LLM ranking preferences.

1192

Interactive Data Lab @idl.uw.edu · 26/11/2024

👋

010

Interactive Data Lab @idl.uw.edu · 25/11/2024

Hi Bluesky! 👋 We’re the Interactive Data Lab at UW. We’ll post about data visualization, analysis, and human-computer interaction research, as well as open source projects. Over the years we’ve been involved with Protovis, D3, Data Wrangler (-> Trifacta), Vega/Vega-Lite, Mosaic, and other projects!

3332