Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Chart Generation on PewResearch and CharXiv Human Study Subset (test)
Loading...
94
Type Correctness
ChartDesign
89.3
91.65
94
96.35
Apr 6, 2026
Type Correctness
Orientation Correctness
Layout Fidelity (Likert 1-5)
Visual Plausibility (Likert 1-5)
Preference: Original
Preference: ChartDesign
Preference: None
Updated 15d ago
Evaluation Results
Method
Method
Links
Type Correctness
Orientation Correctness
Layout Fidelity (Likert 1-5)
Visual Plausibility (Likert 1-5)
Preference: Original
Preference: ChartDesign
Preference: None
ChartDesign
2026.04
94
90
4.4
4.6
25
65
10
Feedback
Search any
task
Search any
task