VeriOCRBench - Factual Dimension (Assisted): leaderboard
Metric: Diagnosis accuracy (%; the 400 trap-injected tasks of the factual verification dimension (fact tampering and vague reference: a stated attribute or value contradicts the visible evidence, or a reference has several plausible targets); the model must reject the task and name the ground-truth trap mechanism; judged by GPT-4o at temperature 0; assisted paradigm: a meta-instruction asks the model to verify the task first). Source: arxiv.org. Saturation forecast: Around 2029. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 397B A17B | 78 |
| 2 | GPT-5 | 73 |
| 3 | Claude Sonnet 4.5 | 67.75 |
| 4 | Qwen 3.5 35B A3B | 64.75 |
| 5 | Qwen 3.5 122B A10B | 63.5 |
| 6 | Llama 4 Maverick | 56 |
| 7 | GPT-4o | 47.25 |
| 8 | InternVL3-38B | 45 |
| 9 | Llama 4 Scout | 37.25 |
| 10 | Gemini 3.1 Pro (Preview) | 7 |
Interactive version: theaggregate.ai/benchmark?slug=veriocrbench-factual-dimension-assisted · How It Works · Data refreshed daily, snapshot 2026-09-26.