PaperAudit-Bench (Fast Mode) - Claude Sonnet 4.5 Errors: leaderboard
Metric: Macro-F1 (%; the per-paper harmonic mean of error coverage and finding precision, averaged over the 46 NeurIPS 2025 oral papers of the benchmark's NeurIPS branch as corrupted by claude-sonnet-4-5-20250929 with 10-20 injected errors of eight types; Fast mode, one global review pass over the whole paper; findings matched to the injected errors by a GPT-5.1 judge; the three text-only detectors read the papers with figures removed). Source: arxiv.org. Saturation forecast: Around September 2027. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 | 47.3 |
| 2 | GPT-5 | 38.7 |
| 3 | GLM-4.6 (Non-reasoning) | 38.1 |
| 4 | Qwen 3 235B A22B 2507 Instruct | 36.2 |
| 5 | Gemini 2.5 Pro | 35.4 |
| 6 | Grok 4 | 35.1 |
| 7 | O4 Mini | 30.7 |
| 8 | Kimi K2 0905 | 27.4 |
| 9 | DeepSeek V3.1 (Non-reasoning) | 26.2 |
Interactive version: theaggregate.ai/benchmark?slug=paperaudit-bench-fast-mode-claude-sonnet-4-5-errors · How It Works · Data refreshed daily, snapshot 2026-09-29.