Trustworthy RAG: An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning in Generative AI Systems
Retrieval-Augmented Generation (RAG) grounds Large Language Model (LLM) outputs in external knowledge, but RAG systems usually trust whatever they retrieve, creating a Security-Reliability Gap: high semantic relevance does not guarantee factual truth. Adversaries exploit this through knowledge poisoning, inserting malicious documents to cause targeted misinformation. We propose an Evaluation Agent, middleware that combines Natural Language Inference (NLI) factual verification, a five-signal poison detector with relevance-weighted aggregation, and a Trust Index T = 0.4 F + 0.35 C + 0.25 (1 - P
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 59%AgentCore-8B →
“Fuzzy title match (0.73): “Trustworthy RAG: An Evaluation Agent for Detecting Misinform” ≈ “AgentCore-8B””
- FuzzySimilar title/name (fuzzy) · 87%SWE-agent/SWE-agent →
“Fuzzy title match (0.94): “Trustworthy RAG: An Evaluation Agent for Detecting Misinform” ≈ “SWE-agent/SWE-agent””
- FuzzySimilar title/name (fuzzy) · 87%zhayujie/CowAgent →
“Fuzzy title match (0.94): “Trustworthy RAG: An Evaluation Agent for Detecting Misinform” ≈ “zhayujie/CowAgent””
- FuzzySimilar title/name (fuzzy) · 84%GoogleCloudPlatform/generative-ai →
“Fuzzy title match (0.92): “Trustworthy RAG: An Evaluation Agent for Detecting Misinform” ≈ “GoogleCloudPlatform/generative-ai””
- FuzzySimilar title/name (fuzzy) · 59%NousResearch/hermes-agent →
“Fuzzy title match (0.73): “Trustworthy RAG: An Evaluation Agent for Detecting Misinform” ≈ “NousResearch/hermes-agent””
- FuzzySimilar title/name (fuzzy) · 59%iflytek/astron-agent →
“Fuzzy title match (0.73): “Trustworthy RAG: An Evaluation Agent for Detecting Misinform” ≈ “iflytek/astron-agent””
- LinkedLinked via arxiv author · 85%Balkrishna Giri →
“Trustworthy RAG: An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning in Generative AI Systems”
- LinkedLinked via arxiv author · 85%Md Toufique Hasan →
“Trustworthy RAG: An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning in Generative AI Systems”