Making AI "reason about space in words" might be backfiring 🧭 Here's a new approach that lets it imagine unseen viewpoints instead. Title: Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models URL: 🧭 Overview This work proposes Imaginative Perception Tokens (IPT) to strengthen spatial reasoning in vision language models (VLMs). Rather than forcing spatial logic through language, it keeps "what could be perceived under a different arrangement" as an intermediate perceptual representation. ❓ Challenges Solved VLMs struggle with spatial reasoning: inferring unobserved viewpoints, reasoning through occluded paths, and integrating partial observations. Prior work pushed this into textual chain-of-thought, but forcing visual reasoning through language alone hit a ceiling. 💡 Methodology & Proposed Approach ・Uses the unified VLM backbone BAGEL, trained with IPT supervision ・Formulates three tasks: Perspective Taking (PET), Path Tracing (PT), Multiview Counting (MVC) ・Builds a ~20,000-example dataset with ground truth, answers, and metrics The core idea is treating the perception itself ("if I moved here, I'd see this") as an intermediate representation. 📊 Experimental Results ・IPT improved Multiview Counting (MVC) accuracy by 3.4% ・Path Tracing (PT) reached performance competitive with closed-source models ・IPT supervision outperformed textual chain-of-thought training ・Conversely, textual CoT substantially degraded spatial reasoning #SpatialReasoning# #MultimodalLLM#

🗺️ Even frontier GPT-5 succeeds on just 14.4% of real-world spatial tasks. A new benchmark goes beyond staring at a static image and exposes how weak AI agents still are at active spatial reasoning. Title: SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks URL: 📝 Overview SpatialWorld measures whether multimodal LLMs can solve tasks by actively exploring 3D environments from a vision-only, egocentric viewpoint. It unifies eight different simulators across indoor, outdoor, and digital-game settings under a shared protocol, and evaluates 15 frontier models on 760 human-annotated tasks. The agent gets no prior map and no reference solution; it has to look, move, and decide on its own. ❓ Challenges Solved Prior spatial-reasoning benchmarks relied on passive evaluation via static VQA or pre-recorded video. That can't capture the interactive spatial understanding the real world demands, where an agent must move its own viewpoint to gather visual evidence and replan on the fly under partial observability. There was a large gap between recognizing a static scene and actually moving through an unfamiliar space to get a task done. 💡 Methodology & Proposed Approach ・The task is framed as a vision-only POMDP (Partially Observable Markov Decision Process) ・The agent receives only a natural-language goal and a single native-resolution egocentric RGB image, with no depth, maps, or semantic metadata ・Actions are issued through a high-level text interface covering navigation, viewpoint control, object interaction, and task completion ・It integrates eight backends: indoor (AI2-THOR, ProcTHOR, VirtualHome), outdoor (CARLA, EmbodiedCity), and digital games (Block3D, Snake3D, Rubik's Cube) ・Success is judged by whether the final terminal state satisfies the goal, not by matching the trajectory, and is validated by human annotators ・Beyond success rate, it measures step efficiency against human reference trajectories to surface inefficient behavior 🎯 Use Cases It offers a unified, fair way to evaluate the spatial abilities of home robots and autonomous agents before real-world deployment. It can systematically diagnose where long-horizon tasks that combine navigation and manipulation break down, serving as a rigorous testbed for improving spatial-reasoning models. 📊 Experimental Results ・Across 15 frontier models, physical-task success was 14.4% for GPT-5, 12.2% for Qwen-3.5-397B, 9.2% for Gemini-3.1-Pro, and 9.2% for Kimi-K2.5 ・On digital games, Gemini-3.1-Pro led at 39.0%, followed by GPT-5 at 36.4% ・By complexity, interaction-only tasks averaged 50.2%, navigation-only dropped to 8.6%, and combined navigation-and-interaction collapsed to just 4.2% ・Models with similar success rates showed very different efficiency scores, revealing heavy reliance on trial-and-error exploration ・Model rankings shifted dramatically across environments, with no single model dominating every category #AIAgents# #SpatialReasoning#

Fable 5.1 and GPT-6 Astra have both now perfected the mensa norway iq test for those who don't know, this test is more about vision and spatial reasoning and they also have an offline test that isn't in the training data, which 5.6 terra had already perfected before astra and fable 5.1 perfected mensa norway

🚀 ZDTaichu5.0-9B is now on ModelScope! 🤖 An on-device multimodal model from TaichuAI. At 9B parameters it runs on a single GPU and brings spatial reasoning, embodied AI and agentic tool use to edge deployment. Qwen3.5-9B backbone + C-RADIOv4-H vision encoder, 128K context, any-resolution image and video input. 🧭 Spatial reasoning: leads the compared 10B-scale open VLMs (Qwen3.5-9B, STEP3-VL-10B, gemma4-8B-E4B) and scores above Gemini 3 Pro, Grok 4 and GPT-5.2 on ViewSpatial, MMSI-Bench and MindCube-tiny 🛠️ Agent: highest among the compared open models on TAU2-Bench, Claw-Eval and IFEval 📄 First-tier results on documents, charts, OCR, visual math and video, with a ready-to-use vLLM branch and Docker image 🧠 Entropy-Gated Adaptive Recurrent Reasoning: extra latent refinement steps go only to the hard tokens

An AI that can turn the right corner doesn't mean it can navigate an entire city. UrbanGround measures exactly that gap, at real scale in Hong Kong. UrbanGround: From Local Perception to Spatial Agency in a Real-Scale City 🏙️ Overview MLLMs have shown impressive local spatial skills — visual recognition, short-range movement, VQA — but whether those skills translate into sustained city-scale action was untested. UrbanGround is a physically-simulated replica of Hong Kong, built from geospatial data (OSM, satellite maps) in Unity, that evaluates MLLM spatial agency through 810 manually-verified task instances arranged across a five-level evaluation ladder. 🔍 The Problem Existing spatial reasoning benchmarks are mostly small-scale, synthetic, and short-range. None test the compounding of local decisions into kilometers-long routes, nor robustness to dynamic changes like weather, road closures, or pedestrian crowds. ⚙️ Methodology A three-layer framework (geospatial / simulation / agent) gives models first-person vision and an interactive map interface. Tasks escalate in five levels: ・Level 1: Visual recognition, orientation, active exploration ・Level 2: Short/long-range and instruction-constrained navigation ・Level 3: Implicit destination inference from description ・Level 4: Multi-task scheduling and route optimization ・Level 5: Dynamic adaptation to closures and pedestrians 📊 Results Visual recognition scores 77–93% — relatively strong. Orientation judgment drops to 23–58%. Short-range navigation success (~70%) collapses to near-zero for long-range tasks. Weather and lighting cut QA accuracy by 5–20 points. Pedestrian collision rates exceed 75% across all models. GPT-5.5 and Kimi-K3 lead overall, but all models share the same failure modes at scale. The core finding: "Local abilities do not compose into sustained exploration." Agents move locally compliant routes but cannot maintain spatial estimates beyond visible scenes or revise plans when routes become invalid. #MLLMs# #EmbodiedAI#

Over the next three days, most of you should start getting GPT-6 Sol, and I already have a few recommendations. First, partially reset the custom instructions you built around GPT-5.6 Sol. Some of them actually make 6 Sol worse. Rebuild them around the new model, update your memories, and tune its style for your own use cases. This makes a surprisingly huge difference, especially in writing. The new checkpoint also behaves differently in formatting. Some bolding disappears, some paragraph breaks change, and markdown can feel slightly less consistent until your instructions are adjusted. As for the actual improvements, the biggest ones I’ve noticed so far are: → much stronger 3D/spatial reasoning → significantly better creative writing and tone control → better continuity across long conversations → much stronger retrieval of old memories and details from months ago → better adaptation to your personal writing style over time Do not treat it like “5.6 Sol, but smarter.” It responds much better when you rebuild your setup around it.

This week on The Information Bottleneck we're talking with Julian Togelius (@togelius ) 🥳🥳🥳 Julian directs the Game Innovation Lab at NYU and co-founded For years he's worked on AI and games from both directions: using AI to generate game content and test games, and using games to actually measure what AI can do. Lately he's been digging into why LLMs that write code so well still fail badly at playing games, and what that gap tells us about spatial reasoning and learning in current models. We'll get into game playing as a benchmark, procedural content generation, open-endedness, and where he thinks the AGI conversation goes wrong. What should we ask him? Drop your questions in the comments

#CREtech2024# in New York City was successfully completed last week! The event was a remarkable gathering of thought leaders, innovators, and industry professionals dedicated to revolutionizing the real estate and construction sectors with cutting-edge technology. #CollovAI# was honored to showcase how our advanced #spatialdesignintelligence# is transforming workflows across real estate and interior design. From enabling dynamic, real-time spatial reasoning to streamlining design proposals for clients, our solutions sparked significant interest and enthusiasm among attendees. Our keynote speech was incredibly inspiring, highlighting the immense potential of AI-driven innovation to reshape the future of spaces. We were excited to see the industry’s strong recognition of our vision and technology. We are eager to expand our network of partners and collaborate with forward-thinking companies ready to embrace the future of spatial design and real estate technology. 🫶Let’s work together to bring smarter, more efficient, and visually stunning solutions to the market!If you’re interested in partnering with us, reach out—we’d love to explore opportunities to drive innovation together! #NYC# #RealEstate# #CollovAI#

👋Introducing the #Meissonic# & #Menotico# model series, a collaborative pre-trained image generation model by Collov Labs! This marks another step forward in our research journey, following our work on #3D# Prior Image Synthesis and D-edit. In collaboration with University of California, Berkeley AI Research and Stanford University , this series features highly efficient image generation models based on the MIM (Masked Image Modeling) architecture. Designed to surpass traditional diffusion models (such as SDXL), these models enable efficient generation of 1024x1024 and 512x512 images even at the edge. The #Meissonic# & #Menotico# series represents a milestone for Collov Labs as we pursue advancements in hashtag#spatialdesignintelligence#. Our goal is to create stunning, efficient text-to-image models trained with minimal data and parameters, paving the way for cost-effective pre-training that delivers enterprise-grade solutions. By collaborating with clients in real estate and home decor, Collov offers customizable, compliant, and cost-controlled pre-training solutions. Through our post-training pipeline, MIM models excel in downstream tasks, including: 🖼 Precision: Perfectly scaled image generation 💡 Visual Memory: Retaining 3D and 2D concepts with fidelity, like furniture, cabinetry, and flooring textures 🚀 Spatial Reasoning: Intelligent spatial arrangement and drag-and-drop editing capabilities We’re thrilled to contribute to the open-source community, sparking discussions on YouTube and Reddit from Japan, Korea, the U.S., India, the Middle East, and the U.K. Check out this YouTube tutorial on deploying these models efficiently at the edge: We invite you to explore and discuss our work! Our code and full paper are now available: 💻 Hugging Face: 📄 Full Paper: