Neo-Classic - Author Attribution (3-Shot, CoT): leaderboard
Stylistic author attribution with three example poems per candidate and a prompt to summarise each style before answering, on a contemporary out-of-sample corpus of metrically verified classical Chinese poems by pseudonymous university poets, from Neo-Classic (Zhang et al., 2026). Accuracy in percent on a 100-question sample; higher is better.
Metric: Accuracy (%; pick the pseudonymous author of a contemporary poem from four candidates, given three example poems per candidate and a prompt to summarise each style first; accuracy on a 100-question sample of the task, as reported in the paper's full results table (accuracy@IF); temperature 0.1, top-p 0.9). Source: arxiv.org. Saturation forecast: Around 2029. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 36 | #111 |
| 2 | DeepSeek V3.2 | 34.3 | #403 |
| 3 | GPT-4o | 34 | #741 |
| 4 | Qwen 3 Max | 34 | #401 |
| 5 | Gemini 2.5 Pro | 28 | #262 |
| 6 | Yi 1.5 34B Chat | 22 | #2513 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=neo-classic-author-attribution-3-shot-cot · How It Works · Data refreshed daily, snapshot 2026-11-12.