# The Consistency Illusion: How Multi-Agent Debate Hides Reasoning Misalignment

arXiv: https://arxiv.org/abs/2606.08457 · HTML: https://mlsanity.com/paper/2606.08457/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 14 of its 32 references that are in the map.
- First posted on arXiv 7 Jun 2026.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)
- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 14 of 14)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Agent Hospital: A Simulacrum of Hospital with Evolvable Medical Agents](https://mlsanity.com/paper/2405.02957/) (2024)
- [Jasper and Stella: distillation of SOTA embedding models](https://mlsanity.com/paper/2412.19048/) (2024)
- [Stop Overvaluing Multi-Agent Debate -- We Must Rethink Evaluation and Embrace Model Heterogeneity](https://mlsanity.com/paper/2502.08788/) (2025)
- [Single-Agent LLMs Outperform Multi-Agent Systems on Multi-Hop Reasoning Under Equal Thinking Token Budgets](https://mlsanity.com/paper/2604.02460/) (2026)
- [MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks](https://mlsanity.com/paper/2505.12371/) (2025)
- [Peacemaker or Troublemaker: How Sycophancy Shapes Multi-Agent Debate](https://mlsanity.com/paper/2509.23055/) (2025)
- [Building Efficient Universal Classifiers with Natural Language Inference](https://mlsanity.com/paper/2312.17543/) (2023)
- [TeamMedAgents: Pareto-Efficient Multi-Agent Medical Reasoning Through Teamwork Theory](https://mlsanity.com/paper/2508.08115/) (2025)

## Built on by (most cited first, 3 of 3)

- [Unanimously Wrong: Certified Abstention from How Medical LLM Consensus Forms](https://mlsanity.com/paper/2610.07570/) (2026)
- [More Reasoning Is Not Always Better: Executing Clinical Guidelines with LLMs and Rules](https://mlsanity.com/paper/2610.15912/) (2026)
- [Beyond Answer Sharing: TrueMAD for Targeted Peer Engagement in Multi-Agent Debate](https://mlsanity.com/paper/2610.18240/) (2026)

## Abstract (arXiv)

Multi-agent LLM systems for medical question answering often treat consensus as a reliability signal: if multiple agents agree on an answer, it is presumed trustworthy. However, answer-level consensus does not entail reasoning-level alignment. We introduce CARA (Cross-Agent Reasoning Alignment), a family of automated metrics that measure whether agents who agree on an answer also agree on the reasoning. Applying CARA to a standard debate system on two medical QA benchmarks, MedQA-USMLE and MedThink-Bench, we identify the consistency illusion: a failure mode where debate reduces detectable contradictions between agents while simultaneously decreasing the semantic similarity of their reasoning chains; agents appear to agree more but reason less consistently. To improve this misalignment, we propose the Grounded Debate Protocol (GDP), a prompt-level intervention that requires agents to commit to named medical facts and take explicit stances on other agents' claims. GDP produces large, consistent alignment improvements, with Cohen's d ranging from +1.43 to +1.99, across two datasets and two backbone models, without adding LLM calls or modifying system architecture. Our results motivate cross-agent reasoning alignment as a quantity to audit alongside accuracy in safety-critical domains.