Gradient-based and instruction-override prompt injections largely fail to survive retrieval and reranking in realistic RAG systems, while only LLM-driven injections remain effective end-to-end, and all attacks are detectable by a lightweight guard.
Core: Corpus-based ranking exploitation via llm manipulation
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
SCI-Defense combines perplexity detection, semantic integrity scoring across four manipulation dimensions, and inter-candidate detection to counter GEO attacks, reporting perfect precision on Amazon product data but domain-limited recall on web passages.
citing papers explorer
-
Can It Reach the Generator? Investigating the Survival of Prompt-Injection Attacks in Realistic RAG Settings
Gradient-based and instruction-override prompt injections largely fail to survive retrieval and reranking in realistic RAG systems, while only LLM-driven injections remain effective end-to-end, and all attacks are detectable by a lightweight guard.
-
SCI-Defense: Defending Manipulation Attacks from Generative Engine Optimization
SCI-Defense combines perplexity detection, semantic integrity scoring across four manipulation dimensions, and inter-candidate detection to counter GEO attacks, reporting perfect precision on Amazon product data but domain-limited recall on web passages.