What Counts as the Same Task? Executable Contracts for Robustness Evaluation of Tool-Using Agents
cs.SE, cs.AI, cs.CR, cs.MA
Submitted: 2026-10-10
Updated: 2026-10-10
Terminology
Sources
- StepShield: When, Not Whether to Intervene on Rogue Agents
- Defending against Adaptive Prompt Injection Attacks via Reasoning-enabled Task Alignment
- Progent: Securing AI Agents with Privilege Control
- The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions
- Text Embeddings by Weakly-Supervised Contrastive Pre-training
- Qwen3 Technical Report
- Qwen2.5 Technical Report
Related papers
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
- PackMonitor: Enabling Zero Package Hallucinations Through Decoding-Time Monitoring
- IntentCoding: Amplifying User Intent in Code Generation
- Incentives and Outcomes in Bug Bounties