Showing cs.CRShow all
2 papers · 1 filter
cs.CR2026
BodhiPromptShield: Pre-Inference Prompt Mediation for Surface-Form Privacy Propagation in LLM Agent Pipelines
Bo Ma, Jinsong Wu, Weiqi Yan
In LLM agent pipelines, prompt privacy risk propagates beyond a single model call: raw user content enters retrieval queries, memory writes, tool arguments, OCR-derived text, and l…
cs.CR2026
REAEDP: Entropy-Calibrated Differentially Private Data Release with Formal Guarantees and Attack-Based Evaluation
Bo Ma, Jinsong Wu, Wei Qi Yan
Sensitive data release is vulnerable to output-side privacy threats such as membership inference, attribute inference, and record linkage. This creates a practical need for release…