Benchmark RadarAI BENCHMARK PROFILE
MV-Bench
MV-Bench evaluates multimodal language models on coordinating multi-view interface construction using Tableau workbooks, converting specifications into executable web interfaces.
- Released
- 2026-07-22
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Multimodal models increasingly generate code from visual designs, but existing evaluations focus on single-chart generation. A dedicated benchmark assesses coordination and data semantics in multi-view interfaces.
Motivation
Multimodal large language models (MLLMs) are increasingly expected to automate visualization development by generating code directly from visual designs.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.