Beyond Division of Labor: How Roles Shape Feedback and Recovery in Multi-Agent Systems
cs.MA
Submitted: 2026-11-02
Updated: 2026-11-02
Terminology
Sources
- Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
- Do Modules Stay in Their Lane? Role Drift in Compound LLM Systems
- Why Do Multi-Agent LLM Systems Fail?
- The Self-Correction Illusion: Role Relabeling Gates Explicit Error Flagging in Large Language Models
- AgentSlimming: Towards Efficient and Cost-Aware Multi-Agent Systems
- Drop the Hierarchy and Roles: How Self-Organizing LLM Agents Outperform Designed Structures
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- "Who Am I, and Who Else Is Here?" Behavioral Differentiation Without Role Assignment in Multi-Agent LLM Systems
- Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models
- Roles with Rails: Contract-Preserving Role Evolution in Multi-Agent Structured Reasoning
- Gemma 4 Technical Report
- CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
- The Llama 3 Herd of Models
- MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
- Large Language Models Cannot Self-Correct Reasoning Yet
- Exact Feedback Is Not Control: Evaluating Text-based Closed-Loop Revision in LLMs
- What Confidence Routing Is Actually Doing: Auditing Routing, Calibration, and Commitment in Multi-Agent Deliberation
- When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs
Related papers
- MultiAgentBench: Evaluating the Collaboration and Competition of LLM agents
- Iterative Negotiation and Oversight: A Case Study in Decentralized Air Traffic Management
- Towards using Large Language Models for Risk Negotiation and Consensus-building in One Health contexts
- A-MapReduce: Executing Wide Search via Agentic MapReduce
- AgentCue: Statistical Summaries for Guiding Decentralized Multi-Agent Self-Organization
- Events as Triggers for Behavioral Diversity in Multi-Agent Reinforcement Learning