Spatial reasoning is a major challenge for the foundation models today, even in simple tasks like arranging objects in 3D space. #CVPR2025 .Introducing LayoutVLM, a differentiable optimization framework that uses VLM to spatially reason about diverse scene layouts from unlabeled

RT @0xSigil: Hiring a founding principal engineer for @extraordinary. $25,000 referral

What do you want to know?.

RT @chrmanning: I’ve joined @aixventureshq as a General Partner, working on investing in deep AI startups. Looking forward to working with….

RT @sanjana__z: 🤖 Household robots are becoming physically viable. But interacting with people in the home requires handling unseen, uncons….

RT @wenlong_huang: How to scale visual affordance learning that is fine-grained, task-conditioned, works in-the-wild, in dynamic envs?. Int….

RT @silasalberti: we trained Kevin-32B = K(ernel D)evin using GRPO on KernelBench. it's to our knowledge the first open model trained using….

RT @sidahuj: 🧑‍🎨 The future of creative tools will look very different. 🧠 Imagine an AI control-centre for orchestrating complex tasks usi….

RT @sidahuj: 🧩 Built an MCP that lets Claude talk directly to Blender. It helps you create beautiful 3D scenes using just prompts!. Here’s….

Huge thanks to the amazing team: @Weiyu_Liu_ (co-lead), Siyi Gu, @dill_pkl , Goutam Bhat, @fedassa , @ManlingLi_ , @nickhaber , @jiajunwu_cs . 🌐Project site: 💻 Code (we plan to open-source everything):. n/n.

Beyond research, consider this: Rockstar Games spends $100M+ and countless human hours meticulously placing 3D assets to create immersive game worlds like GTA. When combined with asset generation models, a model that can spatially reason could automate content creation for

Automated 3D layout generation unlocks richer simulation environments for robotics and embodied AI, enabling:.🔹 More realistic scenes and layouts during training .🔹 Improved generalization for real-world deployment. Consider scene_synthesizer by @clembow, which shares a similar

LayoutVLM outperforms existing methods in our benchmark, where models arrange up to *80* 3D assets given a language instruction and a floor plan. 5/n

The 3D layout optimization landscape is full of local minima—how can we escape them?.🔹 We refine the optimization objectives by validating them against the predicted numerical initialization (code is verifiable!). 🔹 We further finetune our VLM on human-designed 3D scene

Our key idea: Use a VLM to produce two complementary representations and enforce mutual consistency for better spatial reasoning. 🔹 Initialization: predict numerical poses from visually marked multi-view images.🔹 Optimization: generate spatial relations as differentiable

Due to the lack of 3D and dimensional awareness in LLMs, existing methods struggle to generate scenes that are.🔹physically plausible (i.e., no collision).🔹semantically aligned (i.e., objects are placed meaningfully according to the language instruction). 2/n

Claude 3.7 one-shotted this 3D room— everything from layout to geometry and texture. This is always a fun way to test a model’s spatial reasoning ability

RT @YueYangAI: We share Code-Guided Synthetic Data Generation: using LLM-generated code to create multimodal datasets for text-rich images,….

6. Discount the projection by additional risk factors such as AI not getting there, regulation pushback

5. Payback analysis to justify adoption. For example, in fast food, an annual FTE cost is $167,328, including wages ($131,040 at $15/hr), benefits/taxes (20%), and training costs, suggesting humanoids could be cost-competitive with a rental model. "Estimated cost savings of.