One Billion Word
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Language Modeling | One Billion Word (OBW) 100% train set (test) | PPL39.15 | 11 | |
| Language Modeling | One Billion Word (OBW) 1% train set (test) | PPL63.83 | 11 | |
| Language Modeling | One Billion Word corpus 1M sentences | Perplexity71.8 | 5 | |
| Text Generation | One Billion Word (test) | 4-gram JSD0.22 | 2 | |
| Language Modeling | One Billion Word Benchmark (train) | Perplexity36.39 | 2 | |
| Language Generation | One Billion Word 6-gram | JSD0.74 | 2 | |
| Language Generation | One Billion Word 4-gram | JSD0.35 | 2 |
Showing 7 of 7 rows