Summarization Faithfulness on SciGen
37.91SummaCGPT-5
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GPT-5System=GPT-5, Model Class=Large Reasoning Models (LRM), Shots=0-shot2025.12 | 37.91 | 92.73 | |
| o3System=o3, Model Class=Large Reasoning Models (LRM), Shots=0-shot2025.12 | 35.34 | 86.93 | |
| VanillaSystem=Vanilla, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 34.83 | 93.78 | |
| Self-Consistency (SC)System=SC, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 33.19 | 88.79 | |
| Cited Summarization (Cite)System=Cite, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 33.1 | 86.38 | |
| Iterative Refine (IR)System=IR, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 32.63 | 85.05 | |
| Plan-then-Write (Plan)System=Plan, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 30.28 | 83.9 | |
| Decomposition (Deco)System=Deco, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 29.31 | 76.49 | |
| o1System=o1, Model Class=Large Reasoning Models (LRM), Shots=0-shot2025.12 | 29.09 | 97.3 | |
| Chain-of-Thought (CoT)System=COT, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 28.22 | 99.45 | |
| Question-Answer Guided (QAG)System=QAG, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 24.84 | 98.63 | |
| Extract-to-Abstract (E2A)System=E2A, Model Class=LLM (GPT-4.1), Shots=0-shot2025.12 | 24.59 | 97.13 |