zknews

trending4

01▲
A Failure-Mode Benchmark for Polymorphic Sybil Poisoning in RAG arxiv.org
Arxiv benchmark for polymorphic Sybil poisoning in RAG, with forced exposure and a 4-way failure taxonomy instead of yet another toy attack.
arxiv benchmark llm poisoning privacy rag retrieval security
5 pts/deadlock99/2 days ago/2 comments
02▲
Code-Level Cost Function Generation for Spatial Image Steganography Using RAG-Enhanced Large Language Models arxiv.org
LLMs are now writing steganography cost functions, with RAG and a self-evolving knowledge base in the loop.
arxiv llm machine-learning optimization privacy rag steganography
1 pt/pavelennis/1 day ago/2 comments
03▲
Embedding Inference Attack arxiv.org
Can you fingerprint the embedding model from unordered retrieval results? Apparently yes, and the paper tests rerankers and RAG too.
arxiv embeddings information-retrieval llm privacy rag security
3 pts/honestmaj/6 days ago/2 comments
04▲
SMSR: Certified Defence Against Runtime Memory Poisoning in Persistent LLM Agent Systems arxiv.org
SMSR signs agent memory with HMAC and randomized ablation, then proves certified robustness against persistent memory poisoning.
arxiv authentication hash llm rag security signatures
0 pts/hbrown/25 days ago/4 comments