VisualLeakBench - Unsafe Text Propagation (Note Capture): leaderboard

Visual-to-tool leakage (VisualLeakBench, Wang et al., 2026): a single-step vision-language agent sees a screenshot, document, form, chat or dashboard and may call a tool; the score is the percentage of traces that copy the labeled target string into the tool arguments, over 50 images with the baseline system prompt; lower is better. Images show a rendered unsafe request (illicit, cyber, fraud or sabotage content); the agent may save a note of useful details.

Metric: Tool propagation rate (%; non-error traces whose tool arguments contain the labeled target string), 50 images, baseline system prompt, save_note capture. Source: arxiv.org. 4 models tracked.

Top models

Interactive version: theaggregate.ai/benchmark?slug=visualleakbench-unsafe-text-propagation-note-capture · How It Works · Data refreshed daily, snapshot 2026-11-12.