Language Modeling on AG News (test)
52.09PerplexityAR
Evaluation Results
| Method | Links | |
|---|---|---|
| ARzero-shot=true2024.10 | 52.09 | |
| GPT-2Zero-shot=true, Trained on=OpenWebText, Training iterations=1M2025.05 | 52.09 | |
| EDLM-ARzero-shot=true2024.10 | 57.27 | |
| EDLM-coARzero-shot=true2024.10 | 57.94 | |
| EDLM-NCEzero-shot=true2024.10 | 60.02 | |
| MLDMzero-shot=true2024.10 | 61.15 | |
| SEDDzero-shot=true2024.10 | 62.09 | |
| LDDM-MDiffusion Framework=Masked Diffusion, Training Steps=1 million, Training Dataset=OpenWebText2025.10 | 62.55 | |
| VADDZero-shot=true, Trained on=OpenWebText, Training iterations=1M2025.05 | 68 | |
| MDLMZero-shot=true, Trained on=OpenWebText, Training iterations=1M, Reproduced=true2025.05 | 68.57 | |
| MDLMDiffusion Framework=Masked Diffusion, Training Steps=1 million, Training Dataset=OpenWebText, Retrained=true2025.10 | 68.62 | |
| SEDD AbsorbDiffusion Framework=Masked Diffusion, Training Steps=1 million, Training Dataset=OpenWebText, Retrained=true2025.10 | 76.54 | |
| UDLMDiffusion Framework=Uniform Diffusion, Training Steps=1 million, Training Dataset=OpenWebText, Retrained=true2025.10 | 76.81 | |
| LDDM-UDiffusion Framework=Uniform Diffusion, Training Steps=1 million, Training Dataset=OpenWebText2025.10 | 76.81 |