Systematic Review Generation on Systematic Reviews Evaluation Set (test)
79.7ScoreInsightAgent
Evaluation results
| Method | Links | |
|---|---|---|
| InsightAgentBackbone=GPT-4o, Automation Level=Interactive2025.04 | 79.7 | |
| InsightAgentBackbone=Llama 3.3, Automation Level=Interactive2025.04 | 70.2 | |
| InsightAgentBackbone=GPT-4o, Automation Level=Autonomous2025.04 | 62.4 | |
| InsightAgentBackbone=Llama 3.3, Automation Level=Autonomous2025.04 | 60.9 | |
| AutoSurveyBackbone=GPT-4o, Automation Level=Autonomous2025.04 | 54 | |
| ChatCiteBackbone=GPT-4, Automation Level=Autonomous2025.04 | 47.1 |