Doc-PP - Implicit Query Leakage: leaderboard
Metric: Leakage Rate (%; responses revealing the protected value, implicit summary-style queries under a non-disclosure policy, page-image input). Source: arxiv.org. Saturation forecast: Around December 2026. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 (Thinking) | 25.7 |
| 2 | Qwen 3 VL 235B A22B (Thinking) | 40.4 |
| 3 | Gemini 3 Pro (Preview) | 46.8 |
| 4 | Gemini 3 Flash (Preview) (Non-reasoning) | 64.6 |
| 5 | Mistral Large 3 | 76.8 |
| 6 | Qwen 3 VL 235B A22B Instruct | 93.5 |
Interactive version: theaggregate.ai/benchmark?slug=doc-pp-implicit-query-leakage · How It Works · Data refreshed daily, snapshot 2026-09-25.