Hang Li

LightGuard introduces a dual-link architecture that uses a physically confined LiFi channel to securely bootstrap cryptographic session keys, thereby mitigating the risk of key exposure inherent in traditional open-air WiFi communication.

The paper presents an end-to-end system that translates high-level operator intents into low-level, safe routing constraints for LEO mega-constellations, achieving high accuracy and safety guarantees.

The paper introduces Jargon, a novel adversarial framework that exploits the vulnerability of LLMs to context-specific safety boundary blurring, achieving high attack success rates across multiple frontier models.

This paper investigates a novel security vulnerability where imperceptible branding hints can be injected into images and subsequently re-rendered onto new objects by generative AI models, proposing both attack scenarios and a robust mitigation solution.

LymphNode is a novel, post-hoc access control framework that protects Deep Neural Networks (DNNs) from model extraction and inversion attacks by enforcing a default-deny policy and selectively restoring utility only for authorized queries.

The paper introduces ASSEMBLAGE-DEEPHISTORY, a novel, comprehensive binary dataset that unifies cross-compiler builds, historical versions, and vulnerability labels into a single, queryable structure.

The paper introduces a Conflict-aware Penalty (CP) and Statistical Loss (SL) framework to stabilize and balance the training of multimodal sentiment analysis models, achieving state-of-the-art performance.

ESPO is a novel reinforcement learning algorithm that detects trajectory failure in large language models and terminates rollouts early, significantly improving performance on mathematical reasoning benchmarks while reducing computational cost.

DenseSteer is a training-free inference-time framework that improves the math reasoning capabilities of small language models by steering their internal representations toward a 'Dense Reasoning' pattern.

This paper investigates the vulnerability of LLM-based automatic grading systems to prompt injection (PI) attacks, demonstrating that current systems are highly susceptible to manipulation that can lead to unfairly high scores.

This paper introduces Ghostwriter, an attack framework demonstrating that LLMs are highly vulnerable to adopting misleading viewpoints when provided with fabricated, yet credible-looking, evidence.

SentinelRAG introduces a novel watermarking framework that embeds style-consistent, fictitious knowledge entries into RAG databases, allowing for reliable detection of unauthorized redistribution while minimizing impact on legitimate queries.