Some tasks are inferred based on the benchmarks list.
The benchmarks section lists all benchmarks using a given dataset or any of its variants. We use variants to distinguish between results evaluated on slightly different versions of the same dataset. For example, ImageNet 32⨉32 and ImageNet 64⨉64 are variants of the ImageNet dataset.
The ECE dataset (Gui et al., 2016a) is collected from SINA city news and contains 2105 instances. Its document has only one emotion word and one or more emotion causes.