VeriHarm-Bench: Measuring Harmful Success in LLM Coding Agents Verifiably
cs.CL, cs.SE
Submitted: 2026-11-02
Updated: 2026-11-02
Comments: Preprint
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Ctrl-Z: Controlling AI Agents via Resampling
- The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use Agents
- SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
- Coding Agents Are Guessing: Measuring Action-Boundary Violations in Underspecified DevOps Instructions
- Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety
- Overeager Coding Agents: Measuring Out-of-Scope Actions on Benign Tasks
- Kimi K2.5: Visual Agentic Intelligence
- ClashBench: Conflicts Leading Agents to Seize and Harm
- Qwen3 Technical Report
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data