Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Chart Visual Question Answering on CHARTOM misleading visualizations

55.9QA Accuracy

GPT5

14.50825.2543646.746Feb 27, 2025
Updated 3mo ago

Evaluation Results

MethodLinks
2025.02
55.9-
2025.02
53.2-
2025.02
42-
2025.02
39.931.5
2025.02
39.9-
2025.02
36.428.7
2025.02
34.326.6
2025.02
31.523.8
2025.02
30.823.1
2025.02
2821
2025.02
2821
2025.02
27.320.3
2025.02
26.619.6
2025.02
25.918.9
2025.02
25.218.2
2025.02
24.517.5
2025.02
24.517.5
2025.02
22.415.4
2025.02
21.715.4
2025.02
2114.7
2025.02
18.912.6
2025.02
18.912.6
2025.02
16.110.5