利用博弈均衡检测推理步骤真伪，无需训练即可提升多模态模型准确性。
A Nash Equilibrium Framework For Training-Free Multimodal Step Verification
- 将不同判断者视为博弈参与者，通过纳什均衡识别有效推理
- 在6个基准上提升2.4%至5.2%，优于平均打分法
- 适合需要无训练验证的多模态推理场景
多模态大语言模型常生成包含细微错误的推理链，导致答案错误。现有验证方法存在明显局限：学习型评判器需大量标注数据，且跨任务表现不一。而现有无训练方法仅简单平均各来源得分，忽略了关键洞察——当评分不一致时，这种分歧本身恰恰蕴含了推理步骤是否真实的线索。本文提出一种无训练验证方法，将步骤级验证建模为专业评判者间的协作问题。将评判者互动形式化为纳什均衡博弈，共识表示有效步骤，分歧揭示不稳定性。通过闭式解计算均衡得分，实现对分歧的感知过滤与稳定性导向排序。在六个基准上评估，该方法相较基线模型取得2.4%至5.2%的一致提升，并达到与学习型评判器相当的性能，证明跨模态一致性（而非仅平均置信度）可提供无需任务适配的鲁棒验证信号。
原文摘要 · Abstract (English)
Multimodal large language models often generate reasoning chains containing subtle errors that lead to incorrect answers. Current verification approaches have notable limitations. Learned critics need extensive labeled data and show inconsistent performance across different tasks. Meanwhile, existing training-free methods simply average scores from different sources, missing a key insight: when these scores disagree, that disagreement itself carries important information about whether a reasoning step is truly valid or not. We propose a training-free verification approach that treats step-wise verification as a coordination problem among specialized judges. We formalize these judges' interaction as a Nash equilibrium game where agreement signals valid steps while disagreement reveals instability. Our method computes equilibrium scores through a closed-form solution, enabling both disagreement-aware filtering and stability-conscious ranking of reasoning steps. Evaluated across six benchmarks, our approach achieves consistent improvements of 2.4% to 5.2% over baseline models and shows competitive performance against learned critics, demonstrating that cross-modal agreement (not just average confidence) provides robust verification signals without task-specific adaptation.
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。