Peer Review Evaluation on 480-paper subset of 40 papers (test)
47.851Most ConstructiveHuman
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| HumanType=Human-written2026.06 | 47.851 | 31.252 | 42.504 | |
| RevisedType=LLM-generated, Configuration=Revised2026.06 | 28.51 | 45.209 | 27.463 | |
| Revised + RAGType=LLM-generated, Configuration=Revised, RAG=Enabled2026.06 | 12.39 | 11.909 | 14.885 | |
| NormalType=LLM-generated, Configuration=Normal2026.06 | 11.248 | 11.633 | 15.151 |