9 papers
LatentRefusal: Latent-Signal Refusal for Unanswerable Text-to-SQL Queries
Xuancheng Ren, Shijing Hu, Zhihui Lu +2
In LLM-based text-to-SQL systems, unanswerable and underspecified user queries may generate not only incorrect text but also executable programs that yield misleading results or vi…
CEE: An Inference-Time Jailbreak Defense for Embodied Intelligence via Subspace Concept Rotation
Jirui Yang, Zheyu Lin, Zhihui Lu +6
Large language models (LLMs) are widely used for task understanding and action planning in embodied intelligence (EI) systems, but their adoption substantially increases vulnerabil…
InfoDecom: Decomposing Information for Defending Against Privacy Leakage in Split Inference
Ruijun Deng, Zhihui Lu, Qiang Duan
Split inference (SI) enables users to access deep learning (DL) services without directly transmitting raw data. However, recent studies reveal that data reconstruction attacks (DR…
Prefix Probing: Lightweight Harmful Content Detection for Large Language Models
Jirui Yang, Hengqi Guo, Zhihui Lu +6
Large language models often face a three-way trade-off among detection accuracy, inference latency, and deployment cost when used in real-world safety-sensitive applications. This…
Quantifying Privacy Leakage in Split Inference via Fisher-Approximated Shannon Information Analysis
Ruijun Deng, Zhihui Lu, Qiang Duan +1
Split inference (SI) partitions deep neural networks into distributed sub-models, enabling collaborative learning without directly sharing raw data. However, SI remains vulnerable…
Agent Communications toward Agentic AI at Edge -- A Case Study of the Agent2Agent Protocol
Qiang Duan, Zhihui Lu
The current evolution of artificial intelligence introduces a paradigm shift toward agentic AI built upon multi-agent systems (MAS). Agent communications serve as a key to effectiv…