Exfiltration on ACIARENA Medical
58.47UAAgentVerse
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AgentVerseBase LLM=GPT-4o-mini2026.04 | 58.47 | 19.24 | |
| LLM DebateBase LLM=GPT-4o-mini2026.04 | 55.28 | 16.67 | |
| Self ConsistencyBase LLM=GPT-4o-mini2026.04 | 48.62 | 38.49 | |
| AutoGenBase LLM=GPT-4o-mini2026.04 | 41.24 | 29.92 | |
| CAMELBase LLM=GPT-4o-mini2026.04 | 10 | 10.63 |