<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>generative | Awesome Robotics</title><description>Recent interesting robotics papers</description><link>https://broadcastchannel-6vg.pages.dev</link><item><title>Steerable Scene Generation with Post Training and Inference-Time Search- use a diffusion-based framework to generate scalable scenes on procedural data- use a Monte Carlo Tree Search to find physical feasibility and satisfy downstream objective#arxiv #diffusion #generativeX Overview, Project Website</title><link>https://broadcastchannel-6vg.pages.dev/posts/28</link><guid isPermaLink="true">https://broadcastchannel-6vg.pages.dev/posts/28</guid><pubDate>Mon, 12 May 2025 20:35:01 GMT</pubDate><content:encoded>Steerable Scene Generation with Post Training and Inference-Time Search&lt;br /&gt;&lt;br /&gt;- use a diffusion-based framework to generate scalable scenes on procedural data&lt;br /&gt;- use a Monte Carlo Tree Search to find physical feasibility and satisfy downstream objective&lt;br /&gt;&lt;br /&gt;&lt;a href=&quot;/search/%23arxiv&quot;&gt;#arxiv&lt;/a&gt; &lt;a href=&quot;/search/%23diffusion&quot;&gt;#diffusion&lt;/a&gt; &lt;a href=&quot;/search/%23generative&quot;&gt;#generative&lt;/a&gt;&lt;br /&gt;&lt;br /&gt;&lt;a href=&quot;https://fixupx.com/NicholasEPfaff/status/1920921273560977633&quot; target=&quot;_blank&quot;&gt;X Overview&lt;/a&gt;, &lt;a href=&quot;https://steerable-scene-generation.github.io/&quot; target=&quot;_blank&quot;&gt;Project Website&lt;/a&gt;&lt;a href=&quot;https://fixupx.com/NicholasEPfaff/status/1920921273560977633&quot; target=&quot;_blank&quot;&gt;
  
  &lt;div&gt;&lt;i&gt;&lt;b&gt;🧵&lt;/b&gt;&lt;/i&gt; Thread • FixupX&lt;/div&gt;
  &lt;div&gt;
  &lt;i&gt;&lt;/i&gt;
  &lt;div&gt;
    &lt;video src=&quot;https://cdn4.telesco.pe/file/ca0932faf2.mp4?token=E0XedLrnj8Jyht-itaC-QMtebLJVY8ms2zHEEZSRU7N3ohQ_Fja1oJocLlnHEMeRRVac5jbENC276yYnxNw9k9Mkz73q3YOcNCI_CChzzXBih0HiybKvm76HZ6RD6dRCnn7x0ziUY1qDADGdhvs_grcvyLrYn9awjxJWe7Q_oe0Sp56bwzV3O-BYqCRIeOR1KIe3B7zgTjelJiiN9Cx6yj8nybYKsGwhEwpIiymUxMd4SeR7JVE4tzUz3EE5PSr0MPlsHxX2QfE_XTn9VkhnNChgzMCia5fUQ6_ENqOC3RomN9xtkimhNGs1ohrdXWQZi-S3nx4X9E3i6XM3nGGVNA&quot; width=&quot;100%&quot; height=&quot;100%&quot;&gt;&lt;/video&gt;
  &lt;/div&gt;
  &lt;div&gt;&lt;/div&gt;
&lt;time&gt;0:39&lt;/time&gt;
&lt;/div&gt;
  &lt;div&gt;Nicholas Pfaff (@NicholasEPfaff)&lt;/div&gt;
  &lt;div&gt;Want to scale robot data with simulation, but don’t know how to get large numbers of realistic, diverse, and task-relevant scenes?&lt;br /&gt;&lt;br /&gt;Our solution:&lt;br /&gt;➊ Pretrain on broad procedural scene data&lt;br /&gt;➋ Steer generation toward downstream objectives&lt;br /&gt;&lt;br /&gt;&lt;i&gt;&lt;b&gt;🌐&lt;/b&gt;&lt;/i&gt; https://steerable…&lt;/div&gt;
&lt;/a&gt;</content:encoded></item></channel></rss>