Tim Suchanek 推出 Expand Fetch：用视觉模型减少网页 Agent 幻觉
网页 Agent 答错很多时候是把页面读错了，Expand 用视觉模型读原始布局，基准成绩 74% 对比 55%，做 Agent 的可以试试。
Tim Suchanek 团队发布 Expand Fetch，专门解决网页 Agent 因 Markdown 扁平化丢失页面细节导致的误读问题。产品包含两个 System One 模型：视觉模型 Iris 负责保留表格、图标和布局，JSON 模型 X-Ray 提取页面底层数据。在 1,094 个问题、326 个网页的基准测试中，使用 Expand 的 Agent 答对 74%，次优 API 为 55%。每次抓取附带来源回放记录，定价为每 1,000 次抓取 1 美元。
Another great application of System One models. I think much of what we call hallucination in web agents comes from misreading the page. When a page is flattened to text, a small ✗ in a pricing table disappears, and the agent answers incorrectly with full confidence. Expand Fetch reads the rendered page, returns the underlying data as clean JSON, and keeps a replay of exactly what your agent saw. Tim Suchanek @TimSuchanek Introducing Expand Fetch, a new way to scrape and crawl the web. Your agent has never seen the web. No wonder it gets basic questions wrong. Traditional Markdown APIs lose critical details. So we trained two System one models to change that: Iris, a vision model that sees pages like you do, preserving tables, icons, and layouts. X-Ray, a JSON model that uncovers the hidden data behind every page. Every fetch also comes with a Wayback Machine. Every line links back to its source, so you can verify exactly what your agent saw. Three products. One API. On our benchmark of 1,094 questions across 326 challenging web pages, agents using Expand got 74% right. The next best API got 55%. Available today at $1 per 1,000 fetches. Try it at expand.ai Your browser does not support the video tag. 🔗 View on Twitter 🔗 View Quoted Tweet 💬 8 🔄 0 ❤️ 10 👀 1849 📊 8 ⚡
- rohanpaul_ai10-14 16:47原文
