DSPrompt: Dynamic Soft Prompt Defense Against M-RAG Corruption
Multimodal Retrieval Augmented Generation (M-RAG) is increasingly vulnerable to adversarial attacks where malicious data are crafted to produce embeddings that align with benign entries in the vector space, deceiving retrieval and inducing harmful outputs. Existing defenses primarily operate at query time, relying on auxiliary detectors, similarity re-ranking, or feature-consistency checks. However, these approaches suffer from non-trivial inference overhead, generalize poorly to unseen attack strategies, and often assume specific attack distributions. To address this, we propose DSPrompt, a D
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.
- PossiblePossibly related (embedding) · 47%Something keeps turning up in my prompt injection detection logs that I didn't expect. Curious if others doing LLM security work have seen it. [D] →
- FuzzyOverlapping authors or contributors · 62%deepspeedai/DeepSpeed →
“Shared author/contributor keys: lai”
- FuzzyOverlapping authors or contributors · 62%sgl-project/sglang →
“Shared author/contributor keys: zhou”
- FuzzyOverlapping authors or contributors · 62%TauricResearch/TradingAgents →
“Shared author/contributor keys: xiao”
- FuzzyOverlapping authors or contributors · 62%modular/modular →
“Shared author/contributor keys: liu”
- FuzzySimilar title/name (fuzzy) · 59%linshenkx/prompt-optimizer →
“Fuzzy title match (0.73): “DSPrompt: Dynamic Soft Prompt Defense Against M-RAG Corrupti” ≈ “linshenkx/prompt-optimizer””
- LinkedLinked via arxiv author · 85%Chang Liu →
“DSPrompt: Dynamic Soft Prompt Defense Against M-RAG Corruption”
- LinkedLinked via arxiv author · 85%Yuni Lai →
“DSPrompt: Dynamic Soft Prompt Defense Against M-RAG Corruption”