AgentHazard benchmark shows computer-use agents remain highly vulnerable, with attack success rates reaching 73.63% on models like Qwen3-Coder powering Claude Code.
Sg-bench: Evaluating llm safety generalization across diverse tasks and prompt types.AdvancesinNeuralInformationProcessingSystems, 37:123032–123054
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
AgentHazard: A Benchmark for Evaluating Harmful Behavior in Computer-Use Agents
AgentHazard benchmark shows computer-use agents remain highly vulnerable, with attack success rates reaching 73.63% on models like Qwen3-Coder powering Claude Code.