Financial Risk Dashboard Design for Institutional RWA Investors: Visual Hierarchy, Chart Comprehension, and Explainability in FinChart-Bench

Institutional investors who review real-world asset (RWA) credit pools must read dense financial charts while judging default pressure, dilution, liquidity, and liquidation-related exceptions. This paper develops a benchmark-based dashboard design analysis using FinChart-Bench financial chart images and question records. The parsed data release contains 1,202 unique base chart images and 7,019 question records: 2,384 true/false, 2,350 multiple-choice, and 2,285 numeric or short-answer QA records. Each question was matched to its base chart image, classified by chart-reading cue, and mapped conservatively to RWA dashboard panels. Image-level visual features were also extracted from the chart files to estimate the interface burden attributable to visual density, edge structure, colour variation, and mark coverage. The analysis shows that numeric arithmetic is the dominant chart-comprehension demand, covering 3,263 records (46.49%). Comparison, extreme/rank lookup, and trend/direction cues account for another 3,031 records (43.18%). The risk-panel mapping is intentionally cautious: 3,992 records remain general financial context, while 1,345 map to dilution risk, 964 to liquidity risk, 616 to probability of default, and 102 to liquidation anomaly. The findings support a dashboard design principle for institutional RWA review: chart assistance should preserve the original evidence, mark the relevant visual region, attach a concise risk-specific explanation, and keep caveats visible. The contribution is a data-driven interface requirement profile for future VLM-connected dashboards and analyst user testing, rather than a claim of observed analyst performance.