Ming Liu

This study investigates user perceptions of privacy risks associated with GenAI smartphones, finding that users express heightened concerns across the entire data lifecycle and suggest comprehensive, system-level privacy enhancements.

HadAgent introduces a decentralized AI serving system that replaces resource-intensive Proof-of-Work with Proof-of-Inference (PoI) to secure LLM agent operations and achieve fast, verifiable consensus.

The paper introduces MGTEVAL, a comprehensive and extensible platform designed to systematically evaluate the performance, robustness, and efficiency of machine-generated text detectors.

This paper introduces Security Cube, a comprehensive, multi-dimensional framework for evaluating LLM robustness against jailbreak attacks, providing a systematic taxonomy and benchmark analysis of existing attacks and defenses.

The paper introduces DeepTrap, an automated framework that evaluates security vulnerabilities in agentic language models by manipulating their internal execution contexts, demonstrating that task completion does not guarantee safety.

The paper introduces ASPI, a benchmark showing that requiring LLM agents to seek clarification significantly amplifies their vulnerability to prompt injection attacks.

PEMark proposes a novel, non-invasive watermarking scheme that embeds traceability information into API responses by exploiting the permutation redundancy of key-value pair ordering, requiring no modification to existing business systems.

The paper argues that current search agents often verify existing knowledge rather than genuinely searching, and introduces LiveBrowseComp, a new benchmark to measure true evidence-driven discovery.

The paper proposes ESRT, an edge-cloud framework that achieves state-of-the-art, bandwidth-efficient, and privacy-preserving many-to-many speech translation across 45 languages by splitting the model inference.

BORA is an offline-to-online RL framework that enhances dexterous VLA models for real-world robotics by using an action-conditioned critic and a lightweight residual adaptation mechanism to correct execution errors.

The paper introduces Compass, an expert-guided LLM agent framework that successfully extracts and integrates thousands of previously inaccessible marine lead records from vast corpora of scientific papers, creating a major new global database.

The paper introduces AdvCL, a framework that repurposes adversarial perturbations as a geometric control signal to stabilize continual learning in large language models, significantly reducing forgetting and enhancing robustness.

The paper proposes Credit-Attenuated Privileged Feedback (CAPF), a training-time mechanism that uses verifier-side information to guide LLM search agents, significantly improving their performance on complex QA tasks.

This paper introduces Imaginative Perception Tokens (IPT) to improve spatial reasoning in vision language models.