Zi Li

The paper investigates the security risk of extracting knowledge from quantized LLMs deployed on edge devices, showing that structured querying can effectively bypass quantization protections.

This paper introduces FaceTell, a novel side-channel attack system that demonstrates the feasibility of eavesdropping on fine-grained on-screen application activities by analyzing subtle reflections visible on human faces during video conferencing.

The paper introduces Argus, a novel multi-agent framework that reorchestrates Static Application Security Testing (SAST) by integrating LLMs with existing tools to achieve superior, reliable, and cost-effective vulnerability detection.

This paper introduces a novel supply-chain attack that uses model code backdoors to actively steal sensitive secrets from local LLM fine-tuning datasets, bypassing current privacy defenses.

This paper introduces Mobius Injection, a novel, lightweight attack that weaponizes autonomous LLM agents into zombie nodes to launch highly scalable AbO-DDoS attacks by exploiting a vulnerability called Semantic Closure.

The paper introduces Acoustic Interference Attack (AIA), a novel jailbreak method that bypasses Large Audio Language Model (LALM) safety alignments by manipulating the underlying acoustic latent semantics rather than injecting malicious content.

AlphaTransit introduces a novel search-based planning framework that combines Monte Carlo Tree Search (MCTS) with a neural policy-value network to efficiently design high-quality, city-scale bus transit networks.

LongTraceRL addresses long-context reasoning challenges by generating highly challenging training data and introducing a fine-grained rubric reward, significantly improving evidence-grounded reasoning in LLMs.

The paper proposes EAGLE, a novel evidence-aligned multi-agent framework, demonstrating that requiring shared visual evidence among agents is crucial for achieving reliable and trustworthy consensus in multimodal Visual Question Answering (VQA).

The paper introduces SGAP-PPIS, a structure-guided adaptive propagation model that improves protein-protein interaction site prediction by allowing information diffusion to adapt based on a residue's local geometric environment.

This paper introduces CHERRL, a controllable hacking environment for rubric-based reinforcement learning to study and mitigate reward hacking.

This paper presents EurekAgent, an environment-engineered agent system for metric-driven autonomous scientific discovery.