TL;DR Treating spatial reasoning as a classification problem over finite candidates, instead of generating text, speeds up a VLM by up to 5.6x while also improving accuracy. Title: MJev-Spatial: System One Spatial Intelligence via Native Attention Matching URL: Here are the key points. ・🎯 Choices, numeric values, and point coordinates are all unified as classification over a finite candidate set ・🧠 No new prediction head is added - it reuses the model's existing query-key attention projections for scoring ・📐 Point prediction uses a 3x3 grid refined through 3 rounds of re-cropping and re-encoding ・🚀 160.6ms average latency on a B200 GPU, 2.5x faster than the best autoregressive baseline and 5.6x faster than the base model ・📊 65.64% average across 12 benchmarks, beating Qwen3-VL-8B and others, ranking #1# on 7 of them ・🔢 Surprisingly, coarser distance bins actually beat finer ones in accuracy ・🌲 A probability-tree decoding scheme extends the single-point predictor to multi-point prediction with no extra training What's interesting is how this shift away from text generation toward a more "intuitive" classification approach delivers both speed and accuracy at once. #VLM# #SpatialReasoning#

TL;DR Treating spatial reasoning as a classification problem over finite candidates, instead of generating text, speeds up a VLM by up to 5.6x while also improving accuracy. Title: MJev-Spatial: System One Spatial Intelligence via Native Attention Matching URL: Here are the key points. ・🎯 Choices, numeric values, and point coordinates are all unified as classification over a finite candidate set ・🧠 No new prediction head is added - it reuses the model's existing query-key attention projections for scoring ・📐 Point prediction uses a 3x3 grid refined through 3 rounds of re-cropping and re-encoding ・🚀 160.6ms average latency on a B200 GPU, 2.5x faster than the best autoregressive baseline and 5.6x faster than the base model ・📊 65.64% average across 12 benchmarks, beating Qwen3-VL-8B and others, ranking #1# on 7 of them ・🔢 Surprisingly, coarser distance bins actually beat finer ones in accuracy ・🌲 A probability-tree decoding scheme extends the single-point predictor to multi-point prediction with no extra training What's interesting is how this shift away from text generation toward a more "intuitive" classification approach delivers both speed and accuracy at once. #VLM# #SpatialReasoning#

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#

GPT-6 Astra's spatial reasoning is one of its most underrated leaps it's not just recognizing what's on a screen -- it's understanding how objects relate, navigating complex environments, and adapting as things change this is what makes its computer use so phenomenal and such a major breakthrough

Three key model mistake types ➤ Missed grasps and placements, resulting in drops and task materials being pushed out of reach ➤ Poor spatial reasoning, resulting in an incorrect sequence of action and knocking over their own towers ➤ Misjudging success or failure, often continuing after a task reach had reached a passing state

i wouldn't say anthropic is totally crushing openai right now maybe for coding-focused 6.1 sol is incredible and so is the price. yes oai did seem to hit code red but Astra is still unmatched in: frontier math cybersecurity scientific agent workflows computer use spatial reasoning and astra 6.1 is coming soon...

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