修复评估偏差，在 Amazon Bedrock 上选出更优模型
原标题：Fix evaluation bias to unlock better models on Amazon Bedrock
AI 摘要
Property Finder 的工程团队发现，其视觉 AI 评估数据集中 96.8% 的标签直接来自现有模型的预测，而非人工核验，导致评估结果系统性偏向旧模型，任何替代方案都显得表现不佳。他们与 AWS Generative AI Innovation Center 合作，由领域专家重新独立标注 356 张阿联酋房产图片，将单标签分类改为多标签分类，并采用进化式提示优化在 Amazon Bedrock 上迭代选择模型，最终在准确率提升的同时将成本降低 86%。
正文节选
Fix evaluation bias to unlock better models on Amazon Bedrock This post was co-authored with Khaled Thabit, Meriem Boughrara, and Mohamed Desouky from Property Finder. The engineering team at Property Finder, a digital real estate application, discovered that 96.8 percent of their evaluation labels had been inherited directly from a single model’s predictions rather than verified by human experts. Their ground truth was systematically skewed toward the incumbent model, making every alternative a