Extreme Summarization On Gem Xsum
评估指标
Parameters
ROUGE-2
评测结果
各个模型在此基准测试上的表现结果
模型名称 | Parameters | ROUGE-2 | Paper Title | Repository |
|---|---|---|---|---|
| PEGASUS | 568 M | 23.2 | The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics | - |
| PaLM (finetuning)-540B | 540 B | 21.2 | PaLM: Scaling Language Modeling with Pathways | |
| T5-XXL | - | 21.0 | PaLM: Scaling Language Modeling with Pathways | |
| ByT5 | - | - | ByT5: Towards a token-free future with pre-trained byte-to-byte models | |
| PaLM (finetuning)-62B | 62 B | 18.5 | PaLM: Scaling Language Modeling with Pathways | |
| mT5 | - | - | ByT5: Towards a token-free future with pre-trained byte-to-byte models |
0 of 6 row(s) selected.