CommonGen is constructed through a combination of crowdsourced and existing caption corpora, consists of 79k commonsense descriptions over 35k unique concept-sets.
Source: CommonGen: A Constrained Text Generation Challenge for Generative Commonsense Reasoning
Variants: CommonGen
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Text Generation | BART (TextBox 2.0) | TextBox 2.0: A Text Generation … | 2022-12-26 |
| Text Generation | BART | The GEM Benchmark: Natural Language … | 2021-02-02 |
| Text Generation | T5 | The GEM Benchmark: Natural Language … | 2021-02-02 |
| Text Generation | UniLM | CommonGen: A Constrained Text Generation … | 2019-11-09 |
Recent papers with results on this dataset: