Summarization on TriviaQA Generalization
80F1Chain of Summaries (CoS)
Evaluation Results
| Method | Links | |
|---|---|---|
| Chain of Summaries (CoS)Backbone=GPT-4o-mini2025.11 | 80 | |
| Original ContentBackbone=GPT-4o-mini2025.11 | 76 | |
| Zero-shot summaryBackbone=GPT-4o-mini2025.11 | 73 | |
| Chain of DensityBackbone=GPT-4o-mini2025.11 | 73 | |
| PEGASUSBackbone=GPT-4o-mini2025.11 | 72 | |
| BRIOBackbone=GPT-4o-mini2025.11 | 63 | |
| Chain of Summaries (CoS)Backbone=Llama3.2:3B2025.11 | 62 | |
| Chain of Summaries (CoS)Backbone=Qwen2.5:7B2025.11 | 60 | |
| Zero-shot summaryBackbone=Llama3.2:3B2025.11 | 52 | |
| Original ContentBackbone=Llama3.2:3B2025.11 | 51 | |
| PEGASUSBackbone=Llama3.2:3B2025.11 | 50 | |
| Zero-shot summaryBackbone=Qwen2.5:7B2025.11 | 48 | |
| Chain of DensityBackbone=Llama3.2:3B2025.11 | 46 | |
| BRIOBackbone=Llama3.2:3B2025.11 | 45 | |
| PEGASUSBackbone=Qwen2.5:7B2025.11 | 43 | |
| Original ContentBackbone=Qwen2.5:7B2025.11 | 41 | |
| Chain of DensityBackbone=Qwen2.5:7B2025.11 | 36 | |
| BRIOBackbone=Qwen2.5:7B2025.11 | 21 |