11 papers
PPE-Bench: A Benchmark for Evaluating MLLM Unlearning under Private-Public Entanglement
Xianren Zhang, Delvin Ce Zhang, Dongwon Lee +1
Multimodal Large Language Models (MLLMs) have shown strong capabilities, but they may memorize private information from web data, raising privacy concerns. Machine unlearning offer…
When Safe Skills Collide: Measuring Compositional Risk in Agent Skill Ecosystems
Su Wang, Pin Qian, Yihang Chen +6
LLM agents increasingly rely on community-contributed skills that expand an agent's operational capability set. We study a core safety problem in agentic AI systems: whether indivi…
Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents
Minhua Lin, Juncheng Wu, Zijun Wang +14
LLM agents are increasingly deployed as systems built around editable external harnesses, including prompts, skills, memories and tools, that shape task execution without changing…
Moltbook Moderation: Uncovering Hidden Intent Through Multi-Turn Dialogue
Ali Al-Lawati, Nafis Tripto, Abolfazl Ansari +3
The emergence of multi-agent systems introduces novel moderation challenges that extend beyond content filtering. Agents with malicious intent may contribute harmful content that a…
A Functionality-Grounded Benchmark for Evaluating Web Agents in E-commerce Domains
Xianren Zhang, Shreyas Prasad, Di Wang +4
Web agents have shown great promise in performing many tasks on ecommerce website. To assess their capabilities, several benchmarks have been introduced. However, current benchmark…
Query-Efficient Agentic Graph Extraction Attacks on GraphRAG Systems
Shuhua Yang, Jiahao Zhang, Yilong Wang +2
Graph-based retrieval-augmented generation (GraphRAG) systems construct knowledge graphs over document collections to support multi-hop reasoning. While prior work shows that Graph…