用本地开源模型组辩论提升AI对齐能力，效果优于单模型。
Ensemble Debates with Local Large Language Models for AI Alignment
- 构建本地开源模型组进行辩论，通过多视角推理增强对齐性。
- 辩论组在推理深度和论点质量上分别提升19.4%和34.1%。
- 适合关注可复现性与开放评估的AI对齐研究者使用。
随着大语言模型在高风险决策中作用日益突出，其与人类价值观的对齐至关重要。依赖专有API限制了可复现性和广泛参与。我们研究了本地开源模型组辩论是否能提升对齐导向的推理能力。在涵盖15种场景、5种组构型的150场辩论中，组辩论在7分制评分中表现更优（整体得分3.48 vs. 3.13），推理深度提升19.4%，论点质量提升34.1%。改进最显著的是真实性（+1.25分）和人类增强性（+0.80分）。我们公开代码、提示模板和辩论数据集，为基于组辩论的对齐评估提供可访问且可复现的基础。
原文摘要 · Abstract (English)
As large language models (LLMs) take on greater roles in high-stakes decisions, alignment with human values is essential. Reliance on proprietary APIs limits reproducibility and broad participation. We study whether local open-source ensemble debates can improve alignmentoriented reasoning. Across 150 debates spanning 15 scenarios and five ensemble configurations, ensembles outperform single-model baselines on a 7-point rubric (overall: 3.48 vs. 3.13), with the largest gains in reasoning depth (+19.4%) and argument quality (+34.1%). Improvements are strongest for truthfulness (+1.25 points) and human enhancement (+0.80). We provide code, prompts, and a debate data set, providing an accessible and reproducible foundation for ensemble-based alignment evaluation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。