Agent-as-Judge for Factual Summarization of Long Narratives

Benchmark Model Rank Results
long-form-narrative-summarization-on-mensaHierarchically Merging and Agent Refinement#1BERTScore (F1): 60.22ROUGE-1: 31.31ROUGE-2: 8.81ROUGE-L: 18.62
long-form-narrative-summarization-on-moviesumHierarchically Merging and Agent Refinement#1BERTScore (F1): 59.32ROUGE-1: 31.31ROUGE-2: 8.81ROUGE-L: 18.62