| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| NOVELTYBENCH | DeepSeek-14B | Distinct Score52.42 | 31 | 5d ago | |
| MT-Bench | QEMPO-KL | Lexical Diversity Score48.36 | 20 | 3mo ago | |
| INFINITY-EVAL | DeepSeek-14B | Distinct Score39.61 | 16 | 3mo ago | |
| Foundation Benchmark Set | Algorithmic Baseline | Cosine Similarity0.72 | 1 | 2d ago |