| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| CCBench | Qwen3-VL-2B | Accuracy68.88 | 16 | 2mo ago | |
| Korean Cultural Understanding Benchmark (test) | SeaLLM | Abusive Score0.523 | 4 | 5mo ago | |
| Chinese Cultural Understanding Benchmark (test) | SeaLLM | Bias23.7 | 4 | 5mo ago | |
| Arabic Cultural Understanding Benchmark (test) | CultureLLM_Ko | Hate Score57.6 | 3 | 5mo ago |