Happy 4th 🇺🇸!! I have a preview for my next release (assembly101 is massive, so I need to push until next week). But in the meantime, check out the link below to try out the 🚧 work in progress using Rerun and Gradio Links below 👇

Gradio Space: < Conversion Code: <

No jumping, No Running. Workouts at home at any time.🕒🏠 BEST 15 min Beginner Home Workout for Weight Loss 🧘‍♀️🔥

@rerundotio @Gradio Impressive progress! Modular tools like this advance accessibility, transparency, and reproducible AI—exactly where our field needs to be heading. Looking forward to seeing Assembly101 in action soon.

New paper! 📄 Morphology-Adaptive Muscle-Driven Locomotion via Attention Mechanisms If you're at #GECCO2025 🦎 come to the EvoSelf workshop tomorrow Jul 15 to learn more

SceneScript treats 3D reconstruction as a language problem rather than a geometry one. The model watches a video of a room and just learns to write a script for it. It autoregressively spits out text commands like make_wall(...) or make_bbox(...) that define the scene. Stanford's new "Scene Language" paper goes a step further adding CLIP embeddings to capture visual appearance too. The fact that language models already understand spatial relationships well enough to write out scene graphs is pretty wild.

Seek-CAD 🐳 DeepSeek R1-32B model for CAD generation _without_ fine-tuning through in-context learning and self-refinement via VLM (Gemini 2.0) feedback on the R1 CoT. Also a dataset with more ops than DeepCAD- chamfer, revolve, fillets, etc (but unfortunately closed source?)

During training, diffusion models are being taught to be effective denoisers, like Associative Memory systems. At what point do these models stop being denoisers and behaving like data generators? To learn about how these models arise from being Associative Memory systems to generative systems: check out the Modern Methods in Associative Memory tutorial at #ICML2025 in West Ballroom C from 9:30 AM PDT to noon (July 14). The tutorial will be presented by @DimaKrotov, @p_ram_p, @Ben_Hoov from @IBMResearch and @MITIBMLab! See our tutorial paper: See our work which explores the memorization-generalization transition in diffusion models: #icml25 #icml