Ksenia Se

Other important RAG advancements:

• SafeRAG (a benchmark) -> https://huggingface.co/papers/2501.18636
  Establishes a security benchmark revealing how RAG systems are vulnerable to attacks like adversarial data injection, inter-context conflicts, and soft ad poisoning. Evaluates weaknesses in 14 RAG components, emphasizing the need for better filtering and security measures.

• Topic-FlipRAG: Adversarial Opinion Manipulation -> https://huggingface.co/papers/2502.01386
  Demonstrates a two-stage adversarial attack that manipulates RAG-generated opinions on sensitive topics. Alters retrieval rankings and LLM reasoning to subtly flip the stance of generated answers, exposing the difficulty of mitigating semantic-level manipulation.

• Experiments with LLMs on RAG for Closed-Source Simulation Software -> https://huggingface.co/papers/2502.03916
  Tests how RAG can support proprietary software by injecting relevant documentation dynamically. Shows that retrieval helps mitigate hallucinations in closed-source contexts, though some knowledge gaps remain, necessitating further improvements.

• Health-RAG -> https://huggingface.co/papers/2502.04666
  Focuses on medical information retrieval by introducing a three-stage pipeline: retrieve, generate a reference summary (GenText), and re-rank based on factual alignment. Ensures accurate, evidence-backed health answers while mitigating misinformation risks.