PropUQ-MAS: Propagation-Aware Uncertainty Quantification for LLM Multi-Agent Systems
cs.MA, cs.CL
Submitted: 2026-08-22
Updated: 2026-09-01
Code: https://github.com/yaokunliu/PropUQ-MAS
Terminology
Sources
- Position: Safety and Fairness in Agentic AI Depend on Interaction Topology, Not on Model Scale or Alignment
- Training Verifiers to Solve Math Word Problems
- UProp: Investigating the Uncertainty Propagation of LLMs in Multi-Step Agentic Decision-Making
- Gemma 3 Technical Report
- Position: Towards a Responsible LLM-empowered Multi-Agent Systems
- Language Models (Mostly) Know What They Know
- Qwen3 Technical Report
- Cochain: Balancing Insufficient and Excessive Collaboration in LLM Agent Workflows
- Language Agents as Optimizable Graphs
Related papers
- Highway Congestion Reduction through Reinforcement Learning Based Eulerian Headway Control
- You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents
- Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems
- MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimization
- PeroMAS: A Multi-agent System of Perovskite Material Discovery
- StitchCUDA: An Automated Multi-Agents End-to-End GPU Programing Framework with Rubric-based Agentic Reinforcement Learning