Disruption on ACIARENA Code
58.67UAMetaGPT + SANDWICH
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MetaGPT + SANDWICHMAS Platform=MetaGPT, Defense=SANDWICH2026.04 | 58.67 | 0 | |
| MetaGPT + ACI-SENTINELMAS Platform=MetaGPT, Defense=ACI-SENTINEL2026.04 | 44.44 | 7.11 | |
| AgentVerseBase LLM=GPT-4o-mini2026.04 | 35.11 | 45.78 | |
| AgentVerseMAS Platform=AgentVerse, Defense=None2026.04 | 24.89 | 60.44 | |
| AgentVerse + BERT DETECTORMAS Platform=AgentVerse, Defense=BERT DETECTOR2026.04 | 23.56 | 59.78 | |
| AgentVerse + DELIMITERMAS Platform=AgentVerse, Defense=DELIMITER2026.04 | 23.56 | 63.11 | |
| AgentVerse + SANDWICHMAS Platform=AgentVerse, Defense=SANDWICH2026.04 | 23.56 | 60.44 | |
| AgentVerse + G-SAFEGUARDMAS Platform=AgentVerse, Defense=G-SAFEGUARD2026.04 | 20.89 | 74.67 | |
| AgentVerse + AGRAILMAS Platform=AgentVerse, Defense=AGRAIL2026.04 | 20.22 | 65.33 | |
| AgentVerse + ACI-SENTINELMAS Platform=AgentVerse, Defense=ACI-SENTINEL2026.04 | 16.22 | 73.78 | |
| AutoGen + SANDWICHMAS Platform=AutoGen, Defense=SANDWICH2026.04 | 14.22 | 78.67 | |
| Self ConsistencyBase LLM=GPT-4o-mini2026.04 | 13.78 | 76.89 | |
| MetaGPTMAS Platform=MetaGPT, Defense=None2026.04 | 8.89 | 79.56 | |
| CAMELBase LLM=GPT-4o-mini2026.04 | 8.22 | 59.11 | |
| LLM DebateBase LLM=GPT-4o-mini2026.04 | 7.89 | 86.67 | |
| AutoGen + ACI-SENTINELMAS Platform=AutoGen, Defense=ACI-SENTINEL2026.04 | 6.67 | 82.89 | |
| MetaGPT + G-SAFEGUARDMAS Platform=MetaGPT, Defense=G-SAFEGUARD2026.04 | 6.22 | 92.89 | |
| AutoGen + DELIMITERMAS Platform=AutoGen, Defense=DELIMITER2026.04 | 6 | 96.67 | |
| MetaGPT + DELIMITERMAS Platform=MetaGPT, Defense=DELIMITER2026.04 | 6 | 81.78 | |
| MetaGPTBase LLM=GPT-4o-mini2026.04 | 5.33 | 88.89 | |
| AutoGenBase LLM=GPT-4o-mini2026.04 | 4.45 | 90.89 | |
| AutoGenMAS Platform=AutoGen, Defense=None2026.04 | 4.22 | 96.44 | |
| MetaGPT + BERT DETECTORMAS Platform=MetaGPT, Defense=BERT DETECTOR2026.04 | 2.22 | 86.67 | |
| AutoGen + BERT DETECTORMAS Platform=AutoGen, Defense=BERT DETECTOR2026.04 | 2 | 99.78 | |
| AutoGen + AGRAILMAS Platform=AutoGen, Defense=AGRAIL2026.04 | 1.11 | 96.44 | |
| AutoGen + G-SAFEGUARDMAS Platform=AutoGen, Defense=G-SAFEGUARD2026.04 | 0.22 | 96.44 | |
| MetaGPT + AGRAILMAS Platform=MetaGPT, Defense=AGRAIL2026.04 | 0 | 96.89 |