CIBuzzBench - Meaning Explanation (Chinese Prompt): leaderboard
CIBuzzBench (Wang et al., 2026) tests cross-lingual understanding of Chinese internet buzzwords. In Meaning Explanation the model explains the non-literal sense of a buzzword in English under Chinese-language instructions, with the Chinese buzzword as input; human annotators rate 200 sampled explanations per model from 0 to 5 for semantic and pragmatic equivalence to the gold sense, higher is better.
Metric: Human rating (out of 5) (semantic and pragmatic equivalence of the English explanation of the buzzword to the gold annotated sense, scored 0-5 by human annotators on 200 randomly sampled predictions per model; Chinese instructions; 3,001 Chinese internet buzzwords; zero-shot, temperature 0; the Chinese buzzword is the input under both prompt languages). Source: arxiv.org. Saturation forecast: Around February 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 3.9 | #91 |
| 2 | Qwen 3.7 Max | 3.84 | #142 |
| 3 | GLM-5.2 | 3.61 | #137 |
| 4 | GPT-5.5 | 3.13 | #48 |
| 5 | DeepSeek V4 Pro | 3.06 | #181 |
| 6 | Claude Opus 4.8 | 3.03 | #56 |
| 7 | Qwen 3 14B | 2.08 | #1235 |
| 8 | Qwen 3 8B | 1.79 | #1838 |
| 9 | Qwen 3 4B | 1.35 | #2297 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=cibuzzbench-meaning-explanation-chinese-prompt · How It Works · Data refreshed daily, snapshot 2026-11-12.