perplexity_ai· @perplexity_ai · X·· 4 小时前AI 评分
Perplexity沙盒用grep提升金融问答准确率
因为每个沙盒都保存了真实的代码检出,智能体使用 grep 和标准文件工具进行搜索,而非依赖宽泛的语义搜索。 在一项包含 150 个金融文档问题的内部基准测试中,该方法达到了 0.627 的答案准确率,而语义搜索约为 0.37。
Because each sandbox holds a real checkout, agents search it with grep and standard file tools instead of relying on broad semantic searches.
On an internal benchmark of 150 financial-document questions, this reached 0.627 answer accuracy versus about 0.37 for semantic search.
来源:perplexity_ai · x.com