CiteAudit (Generated): leaderboard
Metric: F1 (fraction times 100) for flagging hallucinated references, the hallucinated class positive, on the CiteAudit generated test set (6,086 BibTeX references: 3,586 verified real ones and 2,500 fabricated variants with title, author or metadata errors, each validated by retrieval and human review); each language model classifies every reference as real or hallucinated through its API (tool access is not described); higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 95.5 | #105 |
| 2 | Gemini 3 Pro | 76.9 | #77 |
| 3 | Mixtral 8x7B Instruct (v0.1) | 65.5 | #1049 |
| 4 | Claude Sonnet 4.5 | 59.4 | #138 |
| 5 | Llama 3.3 70B Instruct | 54.8 | #520 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=citeaudit-generated · How It Works · Data refreshed daily, snapshot 2026-10-11.