Han Zhao

The paper introduces WebPII, a novel, large-scale synthetic benchmark for detecting personally identifiable information (PII) in web screenshots, and demonstrates a model (WebRedact) that significantly improves detection accuracy and speed.

The paper proposes a Digital Twin-enabled Simultaneous Learning and Modeling (DT-SLAM) framework to enhance secure communications in UAV-assisted networks against intelligent eavesdropping attacks, achieving significant gains in secure throughput.

This survey provides a comprehensive, structured review of safety research in Embodied AI, analyzing attacks and defenses across the entire embodied pipeline to guide the development of safe, robust, and reliable real-world agents.

The paper introduces ML-Bench, a policy-grounded multilingual safety benchmark, and ML-Guard, a superior guardrail model that enables culturally and legally aligned safety assessment for LLMs across 14 languages.

PRO-CUA introduces a process-reward optimization framework that enables efficient, step-level reinforcement learning for training computer use agents by decoupling environment interaction from policy optimization.

The paper introduces LLMSurgeon, a framework that estimates the domain-level data mixture of a Large Language Model (LLM) using only generated text, thereby providing a post-hoc method to audit the model's 'digital DNA'.

The paper introduces CanLegalRAGBench, a new Canadian legal QA benchmark, and evaluates RAG systems, finding that while open-source models are competitive, automatic evaluations struggle with nuanced legal retrieval and generation.

RogueMerge introduces a unified framework to robustly attack LLM model merging by addressing the challenges of autoregressive decoding, unknown merging configurations, and prompt generalization, significantly outperforming prior methods.

The paper introduces OpAI-Bench, a novel benchmark designed to study how AI authorship signals evolve and accumulate during the progressive co-editing process between humans and AI.