Faithfulness Evaluation on WikiHow (test)
39.06SummaCGPT-5
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-5Model Category=LRM, Prompting=2-shot2025.12 | 39.06 | 74.44 | |
| VanillaBase Model=GPT-4.1, Prompting=2-shot2025.12 | 38.43 | 89.11 | |
| o3Model Category=LRM, Prompting=2-shot2025.12 | 35.1 | 56.23 | |
| Cited Summarization (Cite)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 34.73 | 92.01 | |
| Extract-to-Abstract (E2A)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 34.47 | 92.83 | |
| Question-Answer Guided (QAG)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 34.32 | 85.87 | |
| Chain-of-Thought (COT)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 34.06 | 94.17 | |
| o1Model Category=LRM, Prompting=2-shot2025.12 | 34.06 | 94.17 | |
| Iterative Refine (IR)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 32.76 | 91.91 | |
| Self-Consistency (SC)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 32.52 | 75.7 | |
| Plan-then-Write (Plan)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 30.97 | 86.16 | |
| Decomposition (Deco)Base Model=GPT-4.1, Prompting=2-shot2025.12 | 30.96 | 87.23 |