用模型间一致性判断文字识别对错，无需训练就能自动纠错。
Consensus Entropy: Harnessing Multi-VLM Agreement for Self-Verifying and Self-Improving OCR
- 通过多个模型输出的分歧程度衡量识别可靠性。
- 比现有方法提升42.1%的准确率，且不需额外训练。
- 适合想快速提升OCR质量的研究者和开发者使用。
光学字符识别（OCR）是视觉语言模型（VLMs）和大语言模型（LLM）高质量数据生成的基础。尽管平均识别准确率持续提升，当前顶尖的VLM仍难以检测样本级错误，且缺乏有效的无监督质量控制手段。本文提出共识熵（Consensus Entropy, CE），一种无需训练、与模型无关的可靠性度量方法，通过测量多模型间预测结果的熵来评估输出可信度。核心思想是：正确预测在输出空间趋于一致，错误预测则彼此分散。基于此，我们构建了轻量级的CE-OCR框架，通过集成模型的一致性验证输出、筛选最优结果，并利用自适应路由提升效率。实验表明，CE在质量验证上表现稳健，相比VLM-as-Judge方法，F1分数提升42.1%。CE-OCR在相同开销下持续优于自洽性和单模型基线。值得注意的是，该方法无需训练或监督，可直接集成。代码已开源：https://github.com/Aslan-yulong/consensus-entropy。
原文摘要 · Abstract (English)
Optical Character Recognition (OCR) is fundamental to Vision-Language Models (VLMs) and high-quality data generation for LLM training. Yet, despite progress in average OCR accuracy, state-of-the-art VLMs still struggle with detecting sample-level errors and lack effective unsupervised quality control. We introduce Consensus Entropy (CE), a training-free, model-agnostic metric that estimates output reliability by measuring inter-model agreement entropy. The core insight is that correct predictions converge in output space, while errors diverge. Based on CE, we develop CE-OCR, a lightweight multi-model framework that verifies outputs by ensemble agreement, selects the best outputs, and further improves efficiency through adaptive routing. Experiments demonstrate that CE is robust for quality verification, improving F1 scores by 42.1% over VLM-as-Judge. CE-OCR achieves consistent OCR gains, outperforming self-consistency and single-model baselines at the same cost. Notably, CE requires no training or supervision, enabling plug-and-play integration. Code: https://github.com/Aslan-yulong/consensus-entropy.
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。