List your company

Multi-hop Question Answering on HotpotQA (val) (ASR, ASR-G)

92ASRLogicPoison
49.3660.4371.582.57Apr 3, 2026

Evaluation results

MethodLinks
LogicPoisonLLM=GPT-4o-mini, RAG Framework=Naive RAG2026.049291.6
LogicPoisonLLM=Llama-3.1-8B, RAG Framework=Naive RAG2026.0488.286.6
LogicPoisonLLM=Qwen-3-32B, RAG Framework=Naive RAG2026.048583.8
LogicPoisonLLM=Qwen-3-32B, RAG Framework=GraphRAG2026.0484.292
LogicPoisonLLM=GPT-4o-mini, RAG Framework=GFM-RAG2026.048178
LogicPoisonLLM=Llama-3.1-8B, RAG Framework=GFM-RAG2026.0480.683.8
LogicPoisonLLM=Llama-3.1-8B, RAG Framework=GraphRAG2026.0479.291.2
LogicPoisonLLM=GPT-4o-mini, RAG Framework=GraphRAG2026.0478.492.2
LogicPoisonLLM=Qwen-3-32B, RAG Framework=GFM-RAG2026.0476.676.8
LogicPoisonLLM=Qwen-3-32B, RAG Framework=HippoRAG22026.0476.669.8
LogicPoisonLLM=GPT-4o-mini, RAG Framework=HippoRAG22026.0473.666.2
LogicPoisonLLM=Llama-3.1-8B, RAG Framework=HippoRAG22026.0472.673
PoisonedRAGLLM=Qwen-3-32B, RAG Framework=HippoRAG22026.0471.867.6
PoisonedRAGLLM=GPT-4o-mini, RAG Framework=GFM-RAG2026.0471.670.4
PoisonedRAGLLM=Qwen-3-32B, RAG Framework=GFM-RAG2026.0471.635
PoisonedRAGLLM=Llama-3.1-8B, RAG Framework=HippoRAG22026.0468.463.6
PoisonedRAGLLM=GPT-4o-mini, RAG Framework=HippoRAG22026.046866
PoisonedRAGLLM=GPT-4o-mini, RAG Framework=GraphRAG2026.0466.880
PoisonedRAGLLM=Qwen-3-32B, RAG Framework=GraphRAG2026.0465.880.2
PoisonedRAGLLM=Llama-3.1-8B, RAG Framework=GraphRAG2026.0464.478
PoisonedRAGLLM=GPT-4o-mini, RAG Framework=Naive RAG2026.0461.872.8
PoisonedRAGLLM=Qwen-3-32B, RAG Framework=Naive RAG2026.0456.868.2
PoisonedRAGLLM=Llama-3.1-8B, RAG Framework=GFM-RAG2026.0455.849.4
PoisonedRAGLLM=Llama-3.1-8B, RAG Framework=Naive RAG2026.045158.8