Candidate supply and answer selection shape the value of LLM judging in multi-agent systems
cs.AI, cs.MA
Submitted: 2026-08-26
Updated: 2026-08-30
Comments: 11 figures
Code: https://github.com/YSTLab/mas-reasoning
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Towards a Science of Scaling Agent Systems
- Should we be going MAD? A Look at Multi-Agent Debate Strategies for LLMs
- Minority Sentinel: When to Overturn Majority Voting in Multi-Agent LLM Debates
- Auditing Multi-Agent LLM Reasoning Trees Outperforms Majority Vote and LLM-as-Judge
- AI safety via debate
- When Is Collective Intelligence a Lottery? Multi-Agent Scaling Laws for Memetic Drift in LLMs
- Training Verifiers to Solve Math Word Problems
- Universal Self-Consistency for Large Language Model Generation
- Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
- Evolution without an Oracle: Driving Effective Evolution with LLM Judges
- Constitutional AI: Harmlessness from AI Feedback
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