PaperAudit-Bench (Fast Mode) - Grok 4 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 grok-4 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 November 2027. 10 models tracked.

Top models

#ModelScore
1Gemini 2.5 Pro34.9
2Claude Sonnet 4.533.9
3GPT-533.8
4GLM-4.6 (Non-reasoning)29.6
5Qwen 3 235B A22B 2507 Instruct29.4
6Grok 428.6
7O4 Mini27.7
8Kimi K2 090520.9
9DeepSeek V3.1 (Non-reasoning)19.2

Interactive version: theaggregate.ai/benchmark?slug=paperaudit-bench-fast-mode-grok-4-errors · How It Works · Data refreshed daily, snapshot 2026-09-29.