The Confident Liar: Diagnosing Multi-Agent Debate with Log-Probabilities and LLM-as-Judge
Multi-agent debate systems are typically evaluated only on whether the final answer is correct, overlooking the quality of the intermediate reasoning that debate is designed to produce. This paper studies the relationship between three signals in multi-agent debate: token-level log-probability distributions over reasoning tokens, LLM-as-judge rubric scores assigned to those tokens, and final task accuracy. We examine whether internal confidence signals predict externally evaluated reasoning qual
Record details
Published: 9 June 2026
Source: arXiv
Category: Research
Topics: Healthcare · Agents & autonomy
Retrieved: 14 July 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory
arXiv · 8 June 2026
Making Foresight Actionable: Repurposing Representation Alignment in World Action Models
arXiv · 10 June 2026
The Containment Gap: How Deployed Agentic AI Frameworks Fail Public-Facing Safety Requirements
arXiv · 11 June 2026
Towards Healthy Evolution: Exploring the Role and Mechanisms of Human-Agent Interaction in Self-Evolving Systems
arXiv · 4 June 2026
AgentFairBench: Do LLM Agents Discriminate When They Act?
arXiv · 15 June 2026
Auditing LLM-Governed Social Robots with Culture-Specific Moral Gradients
arXiv · 2 June 2026
How to cite this record
ethics.ai (9 June 2026), “The Confident Liar: Diagnosing Multi-Agent Debate with Log-Probabilities and LLM-as-Judge,” evidence record 1233, https://ethics.ai/record/1233 (originally published by arXiv).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.