List your company

Question Answering on Bio (test)

82.9LLM-Judge ScoreInstructRAG
31.9445.1758.471.63Sep 27, 2025

Evaluation results

MethodLinks
InstructRAGModel=GPT-4o-mini, Poison Position=25, k=502025.0982.9
InstructRAGModel=GPT-4o-mini, Poison Position=50, k=502025.0982.7
Sampling + MISModel=GPT-4o-mini, Attack Position=Benign, k=502025.0981.3
Vanilla RAGModel=GPT-4o-mini, Attack Position=Benign, k=502025.0981.2
AstuteRAGModel=GPT-4o-mini, Poison Position=25, k=502025.0980.8
AstuteRAGModel=GPT-4o-mini, Poison Position=50, k=502025.0980.7
Sampling + MISModel=GPT-4o-mini, Poison Position=50, k=502025.0980.7
Vanilla RAGModel=GPT-4o-mini, Poison Position=1, k=502025.0980.3
Sampling + MISModel=GPT-4o-mini, Poison Position=25, k=502025.0980.2
Sampling + MISModel=GPT-4o-mini, Attack Position=@Pos 50, k=502025.0978.2
AstuteRAGModel=GPT-4o-mini, Attack Position=@Pos 25, k=502025.0978.1
AstuteRAGModel=GPT-4o-mini, Attack Position=@Pos 50, k=502025.0977.1
Sampling + MISModel=GPT-4o-mini, Attack Position=@Pos 25, k=502025.0976.7
InstructRAGModel=GPT-4o-mini, Poison Position=1, k=502025.0976.1
Sampling + MISModel=GPT-4o-mini, Poison Position=1, k=502025.0976.1
Sampling + MISModel=GPT-4o-mini, Attack Position=@Pos 1, k=502025.0975.6
Vanilla RAGModel=GPT-4o-mini, Attack Position=@Pos 1, k=502025.0975.3
InstructRAGModel=GPT-4o-mini, Attack Position=Benign, k=502025.0974.7
InstructRAGModel=GPT-4o-mini, Attack Position=@Pos 25, k=502025.0974.3
Vanilla RAGModel=GPT-4o-mini, Poison Position=25, k=502025.0973.4
AstuteRAGModel=GPT-4o-mini, Attack Position=Benign, k=502025.0973.3
Sampling + MISModel=Llama3.2-3B, Attack Position=Benign, k=502025.0972
AstuteRAGModel=GPT-4o-mini, Poison Position=1, k=502025.0971.8
Sampling + MISModel=Mistral-7B, Attack Position=Benign, k=502025.0970.9
InstructRAGModel=Llama3.2-3B, Attack Position=@Pos 1, k=502025.0969.8
Sampling + MISModel=Mistral-7B, Attack Position=@Pos 25, k=502025.0969.7
Vanilla RAGModel=Mistral-7B, Poison Position=1, k=502025.0969.6
Sampling + MISModel=Mistral-7B, Poison Position=25, k=502025.0969.6
Sampling + MISModel=Llama3.2-3B, Attack Position=@Pos 50, k=502025.0969.5
InstructRAGModel=Llama3.2-3B, Attack Position=Benign, k=502025.0969.3
Vanilla RAGModel=Mistral-7B, Attack Position=Benign, k=502025.0969
Sampling + MISModel=Mistral-7B, Poison Position=50, k=502025.0969
Vanilla RAGModel=Llama3.2-3B, Attack Position=@Pos 1, k=502025.0968.9
Vanilla RAGModel=Llama3.2-3B, Poison Position=1, k=502025.0968.9
Sampling + MISModel=Llama3.2-3B, Poison Position=50, k=502025.0968.9
InstructRAGModel=Mistral-7B, Attack Position=Benign, k=502025.0968.7
Vanilla RAGModel=Mistral-7B, Attack Position=@Pos 1, k=502025.0967.9
Vanilla RAGModel=Llama3.2-3B, Attack Position=Benign, k=502025.0967.9
RobustRAGModel=GPT-4o-mini, Poison Position=50, k=502025.0967.6
RobustRAGModel=GPT-4o-mini, Poison Position=25, k=502025.0967.5
RobustRAGModel=GPT-4o-mini, Attack Position=@Pos 25, k=502025.0967.2
RobustRAGModel=GPT-4o-mini, Poison Position=1, k=502025.0967
Sampling + MISModel=Llama3.2-3B, Attack Position=@Pos 25, k=502025.0966.6
RobustRAGModel=GPT-4o-mini, Attack Position=@Pos 1, k=502025.0966.5
Vanilla RAGModel=GPT-4o-mini, Poison Position=50, k=502025.0966.1
Sampling + MISModel=Mistral-7B, Poison Position=1, k=502025.0965.9
Sampling + MISModel=Llama3.2-3B, Poison Position=25, k=502025.0965.9
InstructRAGModel=GPT-4o-mini, Attack Position=@Pos 50, k=502025.0965.8
RobustRAGModel=GPT-4o-mini, Attack Position=Benign, k=502025.0965.7
RobustRAGModel=GPT-4o-mini, Attack Position=@Pos 50, k=502025.0965.5
Sampling + MISModel=Llama3.2-3B, Poison Position=1, k=502025.0965.3
InstructRAGModel=Mistral-7B, Attack Position=@Pos 1, k=502025.0965.1
Sampling + MISModel=Mistral-7B, Attack Position=@Pos 50, k=502025.0964.2
Sampling + MISModel=Llama3.2-3B, Attack Position=@Pos 1, k=502025.0964.1
InstructRAGModel=Mistral-7B, Poison Position=1, k=502025.0963.8
AstuteRAGModel=GPT-4o-mini, Attack Position=@Pos 1, k=502025.0963.2
RobustRAGModel=Mistral-7B, Attack Position=Benign, k=502025.0961.7
InstructRAGModel=GPT-4o-mini, Attack Position=@Pos 1, k=502025.0961.6
RobustRAGModel=Mistral-7B, Poison Position=25, k=502025.0961.3
AstuteRAGModel=Llama3.2-3B, Attack Position=Benign, k=502025.0961
RobustRAGModel=Mistral-7B, Attack Position=@Pos 50, k=502025.0960.4
InstructRAGModel=Mistral-7B, Poison Position=25, k=502025.0960.4
RobustRAGModel=Mistral-7B, Poison Position=1, k=502025.0960.2
RobustRAGModel=Mistral-7B, Attack Position=@Pos 25, k=502025.0959.8
RobustRAGModel=Mistral-7B, Poison Position=50, k=502025.0959.7
RobustRAGModel=Mistral-7B, Attack Position=@Pos 1, k=502025.0959.4
InstructRAGModel=Llama3.2-3B, Poison Position=1, k=502025.0959.4
AstuteRAGModel=Mistral-7B, Attack Position=Benign, k=502025.0959.2
RobustRAGModel=Llama3.2-3B, Poison Position=50, k=502025.0958.3
RobustRAGModel=Llama3.2-3B, Poison Position=25, k=502025.0958
Sampling + MISModel=Mistral-7B, Attack Position=@Pos 1, k=502025.0957.5
RobustRAGModel=Llama3.2-3B, Attack Position=@Pos 25, k=502025.0957.5
RobustRAGModel=Llama3.2-3B, Attack Position=@Pos 50, k=502025.0957.3
InstructRAGModel=Mistral-7B, Poison Position=50, k=502025.0956.9
RobustRAGModel=Llama3.2-3B, Attack Position=@Pos 1, k=502025.0956.8
AstuteRAGModel=Llama3.2-3B, Attack Position=@Pos 1, k=502025.0956.6
RobustRAGModel=Llama3.2-3B, Attack Position=Benign, k=502025.0956.1
RobustRAGModel=Llama3.2-3B, Poison Position=1, k=502025.0956.1
AstuteRAGModel=Mistral-7B, Attack Position=@Pos 1, k=502025.0955.7
Vanilla RAGModel=Mistral-7B, Poison Position=25, k=502025.0953.5
AstuteRAGModel=Mistral-7B, Poison Position=1, k=502025.0952.6
Vanilla RAGModel=Llama3.2-3B, Poison Position=25, k=502025.0952.5
InstructRAGModel=Llama3.2-3B, Attack Position=@Pos 25, k=502025.0951.9
AstuteRAGModel=Mistral-7B, Attack Position=@Pos 25, k=502025.0951.3
InstructRAGModel=Llama3.2-3B, Poison Position=25, k=502025.0948.2
AstuteRAGModel=Llama3.2-3B, Attack Position=@Pos 25, k=502025.0947.3
AstuteRAGModel=Mistral-7B, Poison Position=25, k=502025.0946.1
Vanilla RAGModel=Mistral-7B, Poison Position=50, k=502025.0945.5
AstuteRAGModel=Mistral-7B, Attack Position=@Pos 50, k=502025.0944.8
AstuteRAGModel=Llama3.2-3B, Poison Position=25, k=502025.0944.7
AstuteRAGModel=Llama3.2-3B, Poison Position=1, k=502025.0944.6
AstuteRAGModel=Mistral-7B, Poison Position=50, k=502025.0944.2
Vanilla RAGModel=GPT-4o-mini, Attack Position=@Pos 25, k=502025.0943.1
Vanilla RAGModel=Mistral-7B, Attack Position=@Pos 25, k=502025.0940.7
AstuteRAGModel=Llama3.2-3B, Poison Position=50, k=502025.0940.7
Vanilla RAGModel=Llama3.2-3B, Attack Position=@Pos 25, k=502025.0940.4
InstructRAGModel=Mistral-7B, Attack Position=@Pos 25, k=502025.0938.4
Vanilla RAGModel=Llama3.2-3B, Poison Position=50, k=502025.0937.9
InstructRAGModel=Llama3.2-3B, Poison Position=50, k=502025.0934.1
AstuteRAGModel=Llama3.2-3B, Attack Position=@Pos 50, k=502025.0933.9