UniLMv2: Pseudo-Masked Language Models for Unified Language Model Pre-Training

Open paper
Benchmark Model Rank Results
question-generation-on-squad11UniLMv2#3BLEU-4: 24.43
abstractive-text-summarization-on-cnn-daily-mailUniLMv2#23ROUGE-1: 43.16ROUGE-2: 20.42ROUGE-L: 40.14

Rank counts only results with a code link.