Query-focused Multi-document Summarization on DIVERSESUMM Relevance
69.72CoverageLLM + DPP-Relevance
Evaluation results
| Method | Links | |
|---|---|---|
| LLM + DPP-RelevanceLLM Model=GPT 4o2025.05 | 69.72 | |
| LLM + DPP-RelevanceLLM Model=Claude2025.05 | 69.37 | |
| LLM + DPPLLM Model=Claude2025.05 | 66.72 | |
| LLM + DPPLLM Model=GPT 4o2025.05 | 66.05 | |
| LLM-Selected KPsLLM Model=GPT 4o2025.05 | 64.43 | |
| Naive LLMLLM Model=GPT 4o2025.05 | 64.1 | |
| LLM-Selected KPsLLM Model=Claude2025.05 | 61.8 | |
| Naive LLMLLM Model=Claude2025.05 | 58.43 | |
| LLM + DPP-RelevanceLLM Model=GPT 3.52025.05 | 54.09 | |
| LLM-Selected KPsLLM Model=GPT 3.52025.05 | 52.92 | |
| LLM + DPPLLM Model=GPT 3.52025.05 | 52.29 | |
| LLM + DPP-RelevanceLLM Model=Llama2025.05 | 45.01 | |
| LLM + DPPLLM Model=Llama2025.05 | 42.24 | |
| Naive LLMLLM Model=GPT 3.52025.05 | 40.8 | |
| LLM-Selected KPsLLM Model=Llama2025.05 | 36.03 | |
| Naive LLMLLM Model=Llama2025.05 | 31.82 |