Text Summarization on BookSum

Metric: BERTScore (F1) (higher is better)

LeaderboardDataset
Loading chart...
#ModelBERTScore (F1)Extra DataPaperDateCode
1NexusSum (Mistral Large)70.7NoNexusSum: Hierarchical LLM Agents for Long-Form ...2025-05-30-
2CachED (BART Large)54.4NoEnd-to-End Long Document Summarization using Gra...2025-01-03-
3SLED (BART Large)52.4NoEnd-to-End Long Document Summarization using Gra...2025-01-03-
4Unlimiformer (BART Base)51.5NoEnd-to-End Long Document Summarization using Gra...2025-01-03-
5Zero-Shot (GPT-4o)47.24NoEnd-to-End Long Document Summarization using Gra...2025-01-03-
6Zero-Shot (Mistral Large)46.42NoNexusSum: Hierarchical LLM Agents for Long-Form ...2025-05-30-