Topic-Aware Contrastive Learning for Abstractive Dialogue Summarization

Open paper
Benchmark Model Rank Results
text-summarization-on-samsum-corpusConDigSum#2ROUGE-1: 54.3ROUGE-2: 29.3ROUGE-L: 45.2BertScoreF1: 54
text-summarization-on-samsum-corpusBART-Large#4ROUGE-1: 52.6ROUGE-2: 27ROUGE-L: 42.1BertScoreF1: 52.1

Rank counts only results with a code link.