2 papers
cs.AI2026
Policy-Invisible Violations in LLM-Based Agents
Jie Wu, Ming Gong
LLM-based agents can execute actions that are syntactically valid, user-sanctioned, and semantically appropriate, yet still violate organizational policy because the facts needed f…
cs.SE2026
TestExplora: Benchmarking LLMs for Proactive Bug Discovery via Repository-Level Test Generation
Steven Liu, Jane Luo, Xin Zhang +7
Given that Large Language Models (LLMs) are increasingly applied to automate software development, comprehensive software assurance spans three distinct goals: regression preventio…