AI TrackerFrontier desk · AEST
Back to news
normaltechai

PlotGround: Grounding Plot Digitization in Real Scientific Figures and Their Source Data

arXiv cs.CL · 6 Oct 2026, 03:57 AEST

PlotGround introduces an automated pipeline for plot digitization benchmarks using real scientific figures and their source data. The system aims to improve the accuracy of recovering plotted values, revealing significant gaps in current models' performance.

Key facts

Read aloud

PlotGround introduces an automated pipeline for plot digitization benchmarks using real scientific figures and their source data. The system aims to improve the accuracy of recovering plotted values, revealing significant gaps in current models' performance. Scientific figures often encode quantitative results that are not readily available in machine-readable form, making accurate plot digitization important for verifying and reusing published findings. We introduce PlotGround, an automated pipeline for building plot digitization benchmarks from real scientific figures and their author-released source data. PlotGround maps figures to source tables, identifies reconstructable panels, and generates quantitative questions with source-grounded reference values. Providing source tables instead of figures raises a coding agent's accuracy from ninety point zero percent to ninety seven point four percent while cutting cost by seventy two percent.

Sources in this cluster