BeSafe-Bench: Unveiling Behavioral Safety Risks of Situated Agents in Functional Environments
This addresses a critical safety gap for deploying autonomous agents in real-world settings, though it is incremental as it builds on existing evaluation methods.
The authors tackled the lack of a comprehensive safety benchmark for autonomous agents by introducing BeSafe-Bench, which revealed that even top agents complete fewer than 40% of tasks safely, with strong performance often leading to severe safety violations.
The rapid evolution of Large Multimodal Models (LMMs) has enabled agents to perform complex digital and physical tasks, yet their deployment as autonomous decision-makers introduces substantial unintentional behavioral safety risks. However, the absence of a comprehensive safety benchmark remains a major bottleneck, as existing evaluations rely on low-fidelity environments, simulated APIs, or narrowly scoped tasks. To address this gap, we present BeSafe-Bench (BSB), a benchmark for exposing behavioral safety risks of situated agents in functional environments, covering four representative domains: Web, Mobile, Embodied VLM, and Embodied VLA. Using functional environments, we construct a diverse instruction space by augmenting tasks with nine categories of safety-critical risks, and adopt a hybrid evaluation framework that combines rule-based checks with LLM-as-a-judge reasoning to assess real environmental impacts. Evaluating 13 popular agents reveals a concerning trend: even the best-performing agent completes fewer than 40% of tasks while fully adhering to safety constraints, and strong task performance frequently coincides with severe safety violations. These findings underscore the urgent need for improved safety alignment before deploying agentic systems in real-world settings.