I came across the term MNLI-(m/mm) in Table 1 of the paper BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. I know what MNLI stands for, i.e. Multi-Genre Natural Language Inference, but I'm just unsure about the -(m/mm) part.
I tried to find some information about this in the paper GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding, but this explained only the basic Multi-Genre Language Inference concept. I assume that the m/mm part was introduced later, but this doesn't make any sense because the BERT paper appeared earlier.
It would be nice if someone knows this or has a paper that explains this.