Source map
The links beside current scores and forecast tests take readers directly to the relevant evidence. This register explains what each evidence block supports here—and what claim would overreach it.
Munzner ↗Brehmer & Munzner ↗Lam et al. ↗
Separate purpose, abstraction, representation, interaction, algorithm, work practice, user performance, and human outcome before selecting an evaluation.
The nine dimensions or their maturity scores; those are this report’s synthesis.
Vega-Lite ↗Data2Vis ↗Draco ↗NL4DV ↗
Declarative interaction, learned generation, constraint-based recommendation, and natural-language specifications had become demonstrable and inspectable.
Reliable framing, real-world delivery, or reader benefit.
Real-chart question answering and multi-stage LLM visualization generation crossed into repeatable public evaluation.
Professional reading reliability, rendered-chart human quality, or downstream outcomes.
Data Formulator 2 ↗nvAgent ↗RealChart2Code ↗Raiven ↗DashArena ↗
Visible transformed data, structured database composition, harder real-data generation, constrained compilation, and open-ended dashboard attempts.
One transferable success rate, production maintenance, or reader outcomes.
A human-readable State–Render–Transition–Constraint plan can expose interaction intent and improve same-pipeline authoring measures.
Independent replication, source truth, representative readers, accessibility, production delivery, maintenance, or autonomy.
Chartography ↗Misleading charts ↗DashboardQA ↗Dashboard2Code ↗POLYCHARTQA ↗
Difficult professional reading, deceptive-chart vulnerability, dashboard navigation, hidden state, and multilingual chart QA remain bottlenecks.
The failure rate on ordinary charts or complete mobile, assistive, authenticated, and reader-tested delivery.
Proactive assistance ↗Touching or Chatting ↗KubeStellar ↗OpenClaw ↗Prism ↗
Bounded comprehension, preference, accessibility, novice production, correction, accepted second-person change, one repeat contributor, one project-authored recovery, one open proposal, and one multi-attempt chain through maintainer recheck can be studied directly.
Representative professional reader benefit, maintenance-authority transfer, or a complete production-to-maintenance episode.