Summarization Faithfulness on Multi-News
38.02SummaCo1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| o1System=o1, Model Class=Large Reasoning Models (LRM), Shots=0-shot2025.12 | 38.02 | 51.92 | |
| Extract-to-Abstract (E2A)System=E2A, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 37.56 | 42.19 | |
| Chain-of-Thought (CoT)System=COT, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 37.07 | 39.78 | |
| Question-Answer Guided (QAG)System=QAG, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 36.44 | 33.61 | |
| GPT-5System=GPT-5, Model Class=Large Reasoning Models (LRM), Shots=0-shot2025.12 | 27.47 | 90.4 | |
| VanillaSystem=Vanilla, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 26.84 | 86.84 | |
| Cited Summarization (Cite)System=Cite, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 25.25 | 87.63 | |
| Iterative Refine (IR)System=IR, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 24.61 | 89.57 | |
| Self-Consistency (SC)System=SC, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 23.64 | 88.77 | |
| Plan-then-Write (Plan)System=Plan, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 23.4 | 87.72 | |
| Decomposition (Deco)System=Deco, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 23.35 | 88.03 | |
| o3System=o3, Model Class=Large Reasoning Models (LRM), Shots=0-shot2025.12 | 23.29 | 88.07 |