Neo-Classic - Ci Tune Pattern (CoT): leaderboard
Ci tune identification among four options with a chain-of-thought prompt, 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 (%; select the Cipai (tune pattern) of a contemporary Ci from part of its text, four options, with a chain-of-thought prompt; 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 February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V3.2 | 71.3 | #403 |
| 2 | Gemini 3 Pro (Preview) | 65 | #111 |
| 3 | Qwen 3 Max | 59 | #401 |
| 4 | GPT-4o | 31 | #741 |
| 5 | Gemini 2.5 Pro | 26 | #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-ci-tune-pattern-cot · How It Works · Data refreshed daily, snapshot 2026-11-12.