5 papers
SEAL: Reinforcing Global Safety in Mixture-of-Experts through Shared Expert ALignment
Qingyu Meng, Yiwei Zha, Jiahuan Pei +3
Mixture-of-Experts (MoE) is a scaling architecture for large language models that activates only a small subset of expert modules per token, enabling massive parameter growth with…
PIIGuard: Mitigating PII Harvesting under Adversarial Sanitization
Mingshuo Liu, Yiwei Zha, Min Chen
Browsing-enabled LLM assistants can fetch webpages and answer contact-seeking queries, creating a practical channel for scraping contact-style personally identifiable information (…
PADBen: A Comprehensive Benchmark for Evaluating AI Text Detectors Against Paraphrase Attacks
Yiwei Zha, Rui Min, Shanu Sushmita
While AI-generated text (AIGT) detectors achieve over 90\% accuracy on direct LLM outputs, they fail catastrophically against iteratively-paraphrased content. We investigate why it…
SMARTFinRAG: Interactive Modularized Financial RAG Benchmark
Yiwei Zha
Financial sectors are rapidly adopting language model technologies, yet evaluating specialized RAG systems in this domain remains challenging. This paper introduces SMARTFinRAG, ad…
MobileAgentBench: An Efficient and User-Friendly Benchmark for Mobile LLM Agents
Luyuan Wang, Yongyu Deng, Yiwei Zha +5
Large language model (LLM)-based mobile agents are increasingly popular due to their capability to interact directly with mobile phone Graphic User Interfaces (GUIs) and their pote…