SciFigPlag-Bench - Reuse Localization: leaderboard
Metric: Paired localization accuracy (%; both the source-side and the suspicious-side boxes of a reused subfigure must reach IoU of at least 0.5; SciFigPlag-Bench pairs of source and suspicious scientific figures (2,582 plagiarized pairs from documented real cases and taxonomy-guided synthesis, 2,541 visually similar negatives); the same task-specific prompts for every model, temperature 0). Source: arxiv.org. Saturation forecast: Around December 2026. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B (IT) | 75.2 |
| 2 | GPT-5.4 | 73.5 |
| 3 | Gemini 3 Flash | 71 |
| 4 | Qwen 3.6 35B A3B | 69 |
| 5 | Claude Sonnet 4.6 | 64.2 |
| 6 | Qwen 3.5 35B A3B | 64.1 |
| 7 | Gemma 4 26B A4B (IT) | 5.8 |
| 8 | Mistral Small 4 | 1.2 |
| 9 | Qwen 3.5 2B | 0.8 |
| 10 | InternVL3.5-8B | 0.3 |
| 11 | Gemma 3 4B (IT) | 0.2 |
| 12 | Pixtral-12B-2409 | 0.2 |
Interactive version: theaggregate.ai/benchmark?slug=scifigplag-bench-reuse-localization · How It Works · Data refreshed daily, snapshot 2026-09-29.