PaperAudit-Bench (Fast Mode) - Doubao Seed 1.6 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 doubao-seed-1-6-251015 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 December 2026. 10 models tracked.

Top models

#ModelScore
1Gemini 2.5 Pro47.6
2Grok 447
3GPT-542.9
4O4 Mini37.8
5Qwen 3 235B A22B 2507 Instruct35.7
6GLM-4.6 (Non-reasoning)33.6
7Claude Sonnet 4.533.2
8DeepSeek V3.1 (Non-reasoning)20.2
9Kimi K2 090516.4

Interactive version: theaggregate.ai/benchmark?slug=paperaudit-bench-fast-mode-doubao-seed-1-6-errors · How It Works · Data refreshed daily, snapshot 2026-09-29.