Metric: BERTScore (F1) (higher is better)
| # | Model↕ | BERTScore (F1)▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | NexusSum (Mistral Large) | 63.53 | No | NexusSum: Hierarchical LLM Agents for Long-Form ... | 2025-05-30 | - |
| 2 | Hierarchically Merging and Agent Refinement | 59.32 | No | Agent-as-Judge for Factual Summarization of Long... | 2025-01-17 | Code |
| 3 | Description Only (LED) | 58.92 | No | MovieSum: An Abstractive Summarization Dataset f... | 2024-08-12 | Code |
| 4 | Two-Stage Heuristic (LED Large) | 58.54 | No | MovieSum: An Abstractive Summarization Dataset f... | 2024-08-12 | Code |
| 5 | Zero-Shot (Mistral Large) | 55.5 | No | NexusSum: Hierarchical LLM Agents for Long-Form ... | 2025-05-30 | - |