od2961/ModeBench download history
od2961/ModeBench is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 630 times (0 in the last 7 days), and 630 times in total. It ranks #27,544 among datasets by monthly downloads.
ModeBench ModeBench measures both correctness and the variety of correct answers produced by language models. Each problem has an executable verifier and a canonical definition of answer identity, so distinct wording alone does not count as a new solution. This release contains three