Benchmark RadarAI BENCHMARK PROFILE
ChartAnno
ChartAnno is a benchmark for evaluating MLLMs on chart annotation generation with 1,200 real-world charts and paired code/instructions. It evaluates annotation quality under different instruction specificity and input settings.
- Released
- 2026-08-04
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Addresses the underexplored task of chart annotation, providing a testbed for comparing MLLM capabilities in generating communicative chart annotations.
Motivation
Multimodal large language models (MLLMs) have made significant progress in chart understanding, generation, and editing, but their ability to annotate existing charts remains underexplored.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.