Text Continuation on Wikipedia 1K random samples (test)
38.9R-1 ScoreTextLDM
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| TextLDMModel Type=Continuous Diffusion, Number of Parameters=768M2026.05 | 38.9 | 8.1 | 17.6 | 82.7 | 10.1 | |
| TextLDMModel Type=Continuous Diffusion, Number of Parameters=328M2026.05 | 27.6 | 6.2 | 16.2 | 81.3 | 10.5 | |
| TextLDMModel Type=Continuous Diffusion, Number of Parameters=114M2026.05 | 27.5 | 5.9 | 15.9 | 81 | 8.9 | |
| GPT-2-largeModel Type=Autoregressive, Number of Parameters=774M2026.05 | 25.1 | 5.7 | 16.1 | 82.2 | 8 | |
| GPT-2-mediumModel Type=Autoregressive, Number of Parameters=355M2026.05 | 25 | 5.3 | 15.7 | 81.8 | 8.16 | |
| Block DiffusionModel Type=Discrete Diffusion, Number of Parameters=170M, block size=162026.05 | 23.9 | 4.2 | 13.8 | 81.6 | 7.23 | |
| GPT-2Model Type=Autoregressive, Number of Parameters=137M2026.05 | 23.3 | 4.7 | 15.1 | 81.6 | 7.89 | |
| Block DiffusionModel Type=Discrete Diffusion, Number of Parameters=170M, block size=82026.05 | 22.9 | 4.1 | 13.4 | 81.3 | 7.7 | |
| Block DiffusionModel Type=Discrete Diffusion, Number of Parameters=170M, block size=42026.05 | 21.9 | 3.8 | 13.1 | 80.9 | 7.3 | |
| SSD-LMModel Type=Continuous Diffusion, Number of Parameters=355M2026.05 | 15.1 | 1.8 | 10.4 | 79.3 | 2.82 |