VeriOCRBench - Contextual Dimension (Proactive): leaderboard
The contextual verification dimension of VeriOCRBench (Zhou et al., 2026): 400 OCR-grounded tasks with the traps domain mismatch and hallucinatory target, where the provided context imposes a wrong domain on the image or refers to a target that does not exist in it. Diagnosis accuracy is the percentage on which the model both rejects the task and names the correct trap mechanism, judged by GPT-4o; higher is better. Proactive setting: the standard prompt, with no instruction to verify the task.
Metric: Diagnosis accuracy (%; the 400 trap-injected tasks of the contextual verification dimension (domain mismatch and hallucinatory target: the provided context imposes a wrong domain on the image or refers to a target that does not exist in it); the model must reject the task and name the ground-truth trap mechanism; judged by GPT-4o at temperature 0; proactive paradigm: no verification instruction). Source: arxiv.org. Saturation forecast: Estimated already saturated. 15 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 86.75 | #244 |
| 2 | Qwen 3.5 397B A17B | 60.75 | #242 |
| 3 | Qwen 3.5 35B A3B | 60.75 | #480 |
| 4 | Gemini 3.1 Pro (Preview) | 56.25 | #91 |
| 5 | Qwen 3.5 122B A10B | 55.25 | #329 |
| 6 | Llama 4 Scout | 46 | #1787 |
| 7 | Llama 4 Maverick | 34.75 | #1105 |
| 8 | InternVL3-38B | 28.75 | #905 |
| 9 | GPT-4o | 22.5 | #741 |
| 10 | GPT-5 | 18.5 | #163 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=veriocrbench-contextual-dimension-proactive · How It Works · Data refreshed daily, snapshot 2026-11-12.