Story Completion on XStoryCloze
67.9AccuracyBYOL-nya
Evaluation results
| Method | Links | |
|---|---|---|
| BYOL-nyaParameters=12B, Pre-training Type=CPT2026.01 | 67.9 | |
| BYOL-nyaParameters=4B, Pre-training Type=CPT2026.01 | 65.98 | |
| Gemma-3Parameters=12B, Pre-training Type=PT2026.01 | 63.53 | |
| Qwen-0.6B-BaseTraining Tokens=N/A, Inference Top-K=22026.01 | 59.43 | |
| Qwen-0.6B-BaseTraining Tokens=N/A, Inference Top-K=N/A2026.01 | 59.23 | |
| DenseTraining Tokens=20B, Inference Top-K=22026.01 | 58.97 | |
| DenseTraining Tokens=20B, Inference Top-K=N/A2026.01 | 58.84 | |
| PHSATraining Tokens=20B, Inference Top-K=N/A2026.01 | 58.5 | |
| PHSATraining Tokens=20B, Inference Top-K=22026.01 | 58.44 | |
| InfLLM v2Training Tokens=20B, Inference Top-K=N/A2026.01 | 57.97 | |
| InfLLM v2Training Tokens=20B, Inference Top-K=22026.01 | 57.97 | |
| BYOL-nyaParameters=1B, Pre-training Type=CPT2026.01 | 55.79 | |
| Gemma-3Parameters=4B, Pre-training Type=PT2026.01 | 54.4 | |
| ApertusParameters=8B, Pre-training Type=25092026.01 | 53.74 | |
| Gemma-3Parameters=1B, Pre-training Type=PT2026.01 | 50.96 | |
| Qwen-3Parameters=14B, Pre-training Type=Base2026.01 | 50.83 | |
| Llama-3.1Parameters=8B2026.01 | 50.43 | |
| Qwen-3Parameters=8B, Pre-training Type=Base2026.01 | 49.11 | |
| Llama-3.2Parameters=1B2026.01 | 48.38 | |
| Qwen-3Parameters=1.7B, Pre-training Type=Base2026.01 | 47.45 |