AI读网页，九成字是白读的
AI 助手每次打开一个网页，往往把整页内容囫囵吞下去，哪怕它只需要其中一句话。这篇做了一个「阅读控制层」，不改模型本身，只在内容进门时把关：先记下这一趟到底要找什么、哪些来源已经看过，再决定去原文哪里取证据、取多少，并给每轮阅读设一个总量上限，超了就退回完整视图兜底。在六个模型、五个场景里，总 token 最多降 92.9%，耗时最多降 89.0%,97% 的设置下任务质量不降反升；网页深搜任务里 token 降三到四成、完成时间降两到四成。它不是你明天能装上的东西，但方向很实在：你抱怨 AI 又慢又贵时，很大一块成本来自它读了太多没用的字。
📄 原文摘要(英文)
Web-facing AI agents repeatedly access webpages, documents, repositories, and tool responses, yet reading interfaces often expose entire objects when only sparse evidence is needed. This over-reading increases token and latency costs and can dilute relevant evidence. We present SparseRead, a reusable reading control layer that governs content admission without modifying the agent model. Its Read Intent Protocol tracks evolving needs and source coverage; Adaptive Reader Backends select how evidence is located and exposed; and the Budgeted Read Gate bounds cumulative raw exposure per episode with faithful full-view fallback. A unified Reader interface supports interchangeable evidence acquisition across heterogeneous resources. Across six models, including Claude Opus 5, and five scenarios, SparseRead reduces total tokens by up to 92.9% and wall time by up to 89.0%, preserving or improving task quality in 97% settings. On Web deep-search tasks, it reduces tokens by 31.0--43.0% and task completion time by 23.0--37.0%. Ablations validate the three designs, while integration with three agent frameworks demonstrates portability.