<?xml version="1.0" encoding="UTF-8"?><oembed><type>video</type><version>1.0</version><html>&lt;iframe src=&quot;https://www.loom.com/embed/29e548f5cad94e5795b211e235ce07ba&quot; frameborder=&quot;0&quot; width=&quot;1920&quot; height=&quot;1440&quot; webkitallowfullscreen mozallowfullscreen allowfullscreen&gt;&lt;/iframe&gt;</html><height>1440</height><width>1920</width><provider_name>Loom</provider_name><provider_url>https://www.loom.com</provider_url><thumbnail_height>1440</thumbnail_height><thumbnail_width>1920</thumbnail_width><thumbnail_url>https://cdn.loom.com/os/loom-media-content/loom/5c795f0f-1411-404e-98f6-2923e16299b9/6ce20af4-a380-4cdc-906b-171ba39b757c/sessions-thumbnails-29e548f5cad94e5795b211e235ce07ba-598ca4d059397424.gif/binary?authToken=eyJraWQiOiJtaWNyb3MtZ3JvdXAvdGRwLW9zLWFyY2hldHlwZS9hOWI5NmI1OWlwM3JxOWRtIiwiYWxnIjoiUlMyNTYifQ.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.ctN7UDTikfDAgHR5QPGY-Lp64l6GbiQPZPD3nyQrcSTAPX910eQd1fDR7YXRAAD2eUg-uUQjjPW6kdtSdVE7OBpv56NZ3nUBJrEjSthTkmdI9jFGHxDs3YU3Hn1bshbP4iMRd8Bcg0098zRPum4m6ERa4Q0eFLBEq7iNxnR7MpFHpBNAFh7m8ZORJECxUpQHU7_ltc5DrF8cAODPq1Vwx4Y7Hv4hhOnG1dG3gQEcSbytBYspPU3dcKL-2k89tmFLZ5vAUdFFCowJt7_tmcFt6HcaMokFplMvPzglIQ5EWgCqrPSfsvF9qUOFS48DlAR6o0vUa_z49yXxczP6V-ul8A&amp;Expires=1790925123&amp;Signature=l3YkAVFvObjXTDs-7uRVjffiGo7hvZXMP3cQ-zD51-yz69aWhwaNx6lKYWpio6cVTzxZwPHCux6dkgU1xPAejoLiV6-G2oxuB%7EvVFi160D9roI1x5T0cnkykr-7a-4CTDrAZ3BpkGTA50Lctd7tInACGxV0fKKzHFV5V7fH1FU1Z-%7EfBOS7RTEu4tAG%7ELxk6yqJiHkzdQXuScShuv4-%7EXHnXAe2kjKGnrSf0T-jXCUH79wcpH5zeVWfFBCnY8MZ%7E0KZQZyVrOtHVydRTqUPlLzhavlSrvd2cPGcY3sKz6A2ITuYO39eDBA%7EgtlRW7W3JfuOw0WGeaVeLsdn%7EGmTiXw__&amp;Key-Pair-Id=KQOSYIR44AIC0</thumbnail_url><duration>320.907</duration><title>ArXiv Agentic RAG:</title><description>This Loom demonstrates a corrective RAG agent that answers questions from five ingested AI research papers and automatically falls back to web search when local context is insufficient. It runs a local-only query to fetch and grade the answer as sufficient from the “Attention is All You Need” research paper. It then shows the web fallback branch by asking a more recent, out-of-corpus technology or AI question, where local retrieval is graded insufficient and the agent searches the real-time web. The author also notes a limitation: scope filtering is currently keyword-based rather than a full classifier.</description></oembed>