When "Must" Becomes "Maybe": Constraint Weakening in LLM Agent Workflows
cs.AI, cs.MA
Submitted: 2026-08-25
Updated: 2026-08-25
Comments: 21 pages, 4 figures
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents
- A General Language Assistant as a Laboratory for Alignment
- ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate
- AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors
- OR-Bench: An Over-Refusal Benchmark for Large Language Models
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- Safe Multi-Agent Behavior Must Be Maintained, Not Merely Asserted: Constraint Drift in LLM-Based Multi-Agent Systems
- Identifying the Risks of LM Agents with an LM-Emulated Sandbox
- Towards Understanding Sycophancy in Language Models
- Simple synthetic data reduces sycophancy in large language models
- StateFlow: Enhancing LLM Task-Solving through State-Driven Workflows
- Agentproof: Static Verification of Agent Workflow Graphs
- AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systems
- AFlow: Automating Agentic Workflow Generation
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection