精选理由
Perplexity AI搞了个新评测WANDR，专门看智能体能不能找到一堆实体并核实每个的具体信息，比单一评分更清楚哪里不行。
WANDR是一个评测基准，用于测试智能体发现大规模实体集并验证每个实体特定事实的能力。该基准提供密集且可解释的评估信号，能揭示智能体在哪个环节失败。其流程还可作为半自动训练数据工厂。由Perplexity AI发布。
原文 · Perplexity
WANDR tests how well agents discover large sets of entities and verify specific facts about each one...
WANDR tests how well agents discover large sets of entities and verify specific facts about each one. It provides a dense, interpretable eval signal that reveals whether an agent fails, and where. The pipeline also doubles as a semi-automated factory for training data. 💬 1 🔄 1 ❤️ 5 👀 1487 📊 2 ⚡