Lei Zhang

This paper provides a comprehensive review of the security vulnerabilities and privacy challenges inherent in the Open Radio Access Network (O-RAN) architecture for the 6G era, systematically categorizing threats and reviewing mitigation strategies.

The paper proposes Mean MAE (MMAE), a novel self-supervised pre-training framework that uses flow mixing and teacher-student distillation to improve encrypted traffic classification by capturing multi-granularity context.

TrafficMoE proposes a Disentangle-Filter-Aggregate (DFA) framework using sparse Mixture-of-Experts to improve encrypted traffic classification by separating header and payload features and adaptively fusing them.

The paper introduces a new benchmark (BGTD) and a multimodal framework (mmTraffic) that enables explainable, evidence-grounded interpretation of encrypted network traffic using LLMs.

The paper introduces a semantics-first verification framework for an implemented Shor oracle for ECDLP in Qrisp, demonstrating that even seemingly correct implementations can fail due to subtle control law violations.

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 analyzes observation masking in long-horizon search agents, finding that its effectiveness depends on a complex interaction between the model's capacity and the retriever's strength, exhibiting an inverted-U shaped gain.

TunerDiT introduces a training-free progressive steering method to enhance multi-event video generation using Diffusion Transformers, achieving state-of-the-art performance by explicitly managing event boundaries and cross-event semantics.

The paper proposes a novel hybrid CNN-CodeBERT framework for three-class credential leakage detection, significantly improving accuracy by explicitly distinguishing genuine secrets from weak or placeholder credentials.

The paper introduces a hybrid CNN-CodeBERT framework for three-class credential leakage detection, significantly improving accuracy by explicitly distinguishing genuine secrets from non-secret placeholders.

The paper proposes a framework to harvest unused computation resources on AI chips for general-purpose tasks using neural architecture search and approximation techniques.

This paper proposes ERA, a framework for efficient multimodal large language models using entropy-guided visual token pruning, rectified attention, and bias-aware token recycling.

This paper introduces Byte-Prefix Marginalization (BPM), a method for consolidating open-weight language models through on-policy distillation while preserving teacher probability mass and maintaining a shared byte space.

The paper proposes TRIDENT, a framework to restore a source speaker's identity from converted audio using a three-pronged architecture.