Zhe Li

GasLiteAA proposes optimizing the ERC-4337 standard by offloading gas sponsorship logic to Trusted Execution Environments (TEE), significantly reducing on-chain gas costs while maintaining security and verifiability.

ClawGuard is a novel runtime security framework that deterministically enforces user-confirmed rules at tool-call boundaries to protect LLM agents from indirect prompt injection.

SafeHarbor is a novel, hierarchical memory-augmented framework that establishes context-aware decision boundaries for LLM agents, achieving state-of-the-art safety while minimizing over-refusal.

OrchJail introduces an orchestration-guided fuzzing framework to systematically jailbreak tool-calling text-to-image agents by exploiting unsafe multi-step tool-orchestration patterns.

This paper presents the first systematic study of membership inference attacks (MIAs) against Vision-Language-Action (VLA) models, demonstrating that these models are highly vulnerable to privacy breaches even when only observing generated actions.

DCVD proposes a dual-channel cross-modal fusion framework that jointly detects software vulnerabilities and precisely localizes the vulnerable lines, outperforming existing state-of-the-art methods.

The paper introduces LITMUS, a novel benchmark that rigorously tests LLM agents for dangerous, physical-layer behavioral jailbreaks in real OS environments, revealing that current agents frequently execute high-risk operations despite safety guardrails.

The paper introduces a new security benchmark and framework to defend LLM agents against 'cognitive poisoning,' where malicious tools build trust through benign feedback before executing a harmful final action.

The paper proposes a unified framework to evaluate how different types of memory transfer benefit multi-trajectory inference for tool-use LLM agents, finding that the optimal memory method depends critically on the underlying inference strategy.

SANA-Streaming introduces a novel, efficient framework that enables real-time, high-resolution streaming video-to-video editing by combining a hybrid diffusion transformer with specialized training and hardware co-design.

This paper systematically evaluates how LLMs uncritically adapt to potentially dangerous user prompts related to eating disorders, finding that specific linguistic cues significantly increase the likelihood of unsafe responses.

The paper introduces GigaSpeechBench, a comprehensive multilingual and multidimensional ASR & AST benchmark with 680 hours of human-annotated speech, featuring 12 low-resource languages, 6 Chinese dialects, 6 English accents, dense terminology, older adult and child speech, and human-annotated translations.

The paper introduces ELSA3D, a unified 3D model that uses elastic semantic anchoring to improve interaction between text and 3D representations, achieving state-of-the-art performance with reduced FLOPs and inference latency.

The paper introduces UAV-DualCog, a benchmark for evaluating multimodal large language models in UAV scenarios for joint self-state and environment-state reasoning.

RAMP is a method for improving click-through rate and conversion rate prediction accuracy in privacy-constrained settings by using a personalized pathway, a non-personalized pathway, and a prediction-alignment architecture.

This paper examines how fine-tuning large audio-language models affects the semantics, decoder accessibility, and temporal output alignment of native audio-token states using temporal audio grounding.