✨SpaceTools: Tool-Augmented Spatial Reasoning via Double Interactive RL

📝 Summary:
SpaceTools introduces Double Interactive Reinforcement Learning DIRL. This two-phase RL framework enables Vision Language Models to coordinate multiple tools for precise spatial reasoning, achieving state-of-the-art performance on benchmarks and real-world robot tasks.

🔹 Publication Date: Published on Dec 3

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.04069
• PDF: https://arxiv.org/pdf/2512.04069
• Project Page: https://spacetools.github.io/
• Github: https://spacetools.github.io/

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#ReinforcementLearning #VisionLanguageModels #Robotics #SpatialReasoning #AI

✨Learning to Reason in 4D: Dynamic Spatial Understanding for Vision Language Models

📝 Summary:
DSR Suite improves vision language models weak dynamic spatial reasoning. It creates 4D training data from videos using an automated pipeline and integrates geometric priors via a Geometry Selection Module. This significantly enhances VLM dynamic spatial reasoning capability while maintaining gen...

🔹 Publication Date: Published on Dec 23

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.20557
• PDF: https://arxiv.org/pdf/2512.20557

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VisionLanguageModels #SpatialReasoning #4D #ComputerVision #AIResearch

✨CoV: Chain-of-View Prompting for Spatial Reasoning

📝 Summary:
Chain-of-View CoV prompting enhances spatial reasoning in 3D embodied question answering for vision-language models. It actively explores environments by selecting question-aligned views and iteratively adjusting camera positions to gather context, leading to significant performance gains across ...

🔹 Publication Date: Published on Jan 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.05172
• PDF: https://arxiv.org/pdf/2601.05172

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#SpatialReasoning #VisionLanguageModels #EmbodiedAI #Prompting #AI

✨CoV: Chain-of-View Prompting for Spatial Reasoning

📝 Summary:
Chain-of-View CoV prompting helps vision-language models improve spatial reasoning in 3D embodied question answering. It actively selects question-aligned views and iteratively adjusts camera positions to gather context, significantly boosting performance without additional training.

🔹 Publication Date: Published on Jan 8

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.05172
• PDF: https://arxiv.org/pdf/2601.05172
• Github: https://github.com/ziplab/CoV

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#SpatialReasoning #VisionLanguageModels #PromptEngineering #EmbodiedAI #AIResearch

✨Everything in Its Place: Benchmarking Spatial Intelligence of Text-to-Image Models

📝 Summary:
Text-to-image models struggle with complex spatial reasoning due to sparse prompts. This paper introduces SpatialGenEval, a new benchmark with dense prompts, showing models struggle with higher-order spatial tasks. A new dataset, SpatialT2I, helps fine-tune models for significant performance gain...

🔹 Publication Date: Published on Jan 28

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2601.20354
• PDF: https://arxiv.org/pdf/2601.20354
• Github: https://github.com/AMAP-ML/SpatialGenEval

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#TextToImage #SpatialReasoning #GenerativeAI #ComputerVision #AIResearch

✨SpatiaLab: Can Vision-Language Models Perform Spatial Reasoning in the Wild?

📝 Summary:
SpatiaLab introduces a comprehensive benchmark to evaluate vision language model spatial reasoning in realistic scenarios. Experiments show a significant performance gap between current models and humans, revealing major limitations in tasks like depth and 3D geometry. This highlights challenges ...

🔹 Publication Date: Published on Feb 3

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2602.03916
• PDF: https://arxiv.org/pdf/2602.03916
• Project Page: https://spatialab-reasoning.github.io/
• Github: https://github.com/SpatiaLab-Reasoning/SpatiaLab

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#VisionLanguageModels #SpatialReasoning #ComputerVision #AIResearch #DeepLearning

✨Unleashing Spatial Reasoning in Multimodal Large Language Models via Textual Representation Guided Reasoning

📝 Summary:
TRACE is a prompting method that enables MLLMs to perform 3D spatial reasoning by generating text-based representations of video environments. This improves spatial question answering and consistently outperforms prior strategies.

🔹 Publication Date: Published on Mar 24

🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.23404
• PDF: https://arxiv.org/pdf/2603.23404

==================================

For more data science resources:
✓ https://xn--r1a.website/DataScienceT

#SpatialReasoning #MLLMs #AI #PromptEngineering #ComputerVision

🔥 MolmoAct2: Action Reasoning Models for Real-world Deployment

💡 The paper presents MolmoAct2, an open action reasoning model for robotics that improves upon previous systems in several ways. Current vision-language-action models aim to provide a single generalist controller for robots, but they have limitations, such as being closed, requiring expensive hardware, or having high latency. MolmoAct2 addresses these issues by introducing several new components, including a specialized vision-language-model backbone called MolmoER, which is trained on a large corpus of data and is designed for spatial and embodied reasoning. The model also includes three new datasets, including the largest open bimanual dataset to date, and an open-weight action tokenizer called OpenFAST. The architecture of the model has been redesigned to include a continuous-action expert and an adaptive-depth reasoning variant called MolmoThink, which reduces latency by only re-predicting depth tokens for scene regions that change between timesteps. The results of the paper show that MolmoAct2 outperforms strong baselines in several simulation and real-world benchmarks, and the model weights, training code, and training data are released for use by others. Overall, MolmoAct2 is a fully open action reasoning model that is designed for practical deployment and advances the state of the art in robotics.

📅 Published on May 4

🔗 Links:
• arXiv: https://arxiv.org/abs/2605.02881
• PDF: https://arxiv.org/pdf/2605.02881
• Project Page: https://allenai.org/blog/molmoact2
• GitHub: https://github.com/allenai/molmoact2 ⭐ 90

🤖 Models citing this paper:
• https://huggingface.co/allenai/MolmoAct2
• https://huggingface.co/allenai/MolmoAct2-SO100_101
• https://huggingface.co/allenai/Molmo2-ER

📊 Datasets citing this paper:
• https://huggingface.co/datasets/allenai/13122025-tool-04
• https://huggingface.co/datasets/allenai/13122025-cut-10
• https://huggingface.co/datasets/allenai/28112025-yam-01

🚀 Spaces citing this paper:
• https://huggingface.co/spaces/allenai/dataset-stats
• https://huggingface.co/spaces/allenai/lerobot-visualizer-v3

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#RoboticsActionReasoning #VisionLanguageModels #EmbodiedAI #BimanualRobotics #SpatialReasoning

🔥 SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning

💡 The paper introduces SpatialClaw, a training-free framework that enables flexible and stateful spatial reasoning in vision-language models. The problem addressed is the limitation of current spatial agents in performing open-ended spatial reasoning tasks, which is due to the design of the action interface that invokes specialist perception modules. Existing spatial agents use either single-pass code execution or a structured tool-call interface, both of which offer limited flexibility for complex 3D/4D spatial reasoning.

The proposed SpatialClaw framework uses code as the action interface, allowing a vision-language model-backed agent to write executable code conditioned on prior outputs. This approach enables the agent to flexibly compose and manipulate perception results and adapt its analysis to intermediate text and visual observations. SpatialClaw maintains a stateful Python kernel pre-loaded with input frames and a suite of perception and geometry primitives.

The results show that SpatialClaw achieves superior performance across diverse 3D/4D spatial reasoning tasks, with an average accuracy of 59.9% across 20 benchmarks. This represents a significant improvement of 11.2 points over the recent spatial agent, with consistent gains across six vision-language model backbones from two model families, without any benchmark- or model-specific adaptation. The paper's contribution is the introduction of a flexible and effective framework for spatial reasoning that can be applied to a wide range of tasks without requiring training or adaptation.

📅 Published on Jun 11

🔗 Links:
• GitHub: https://github.com/huggingface
• arXiv: https://arxiv.org/abs/2606.13673
• PDF: https://arxiv.org/pdf/2606.13673
• Project Page: https://spatialclaw.github.io/

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#SpatialReasoning #VisionLanguageModels #AgenticInterfaces #SpatialArtificialIntelligence #CodeBasedActionInterfaces

🔥 S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence

💡 The paper introduces S-Agent, a spatial reasoning framework that enhances visual language models to enable continuous 3D world understanding from multi-view imagery. The problem addressed is that existing visual language models and tool-augmented agents are limited to static and stateless inference from isolated visual observations, which is insufficient for real-world spatial intelligence.

The S-Agent method involves formulating spatial reasoning as spatio-temporal evidence accumulation, rather than isolated frame-level prediction. This is achieved by casting the visual language model as a semantic planner that decides what evidence is needed, while a hierarchy of spatial tools and experts grounds objects in 2D, lifts them into 3D geometric evidence, and aggregates this evidence into high-level spatial knowledge. The framework also includes a temporal memory mechanism, comprising scene memory and agent memory, which enables evidence integration across frames and reasoning steps.

The results show that S-Agent consistently improves both open-source and closed-source visual language models in a training-free manner. Additionally, supervised fine-tuning on S-Agent-generated spatial trajectories yields S-Agent-8B, a compact spatial agent that significantly surpasses similar-scale baselines and performs comparably to advanced closed-source models. The comprehensive experiments on multi-view and video spatial reasoning benchmarks demonstrate the effectiveness of the S-Agent framework in enhancing spatial intelligence. Overall, the paper contributes a novel spatial tool-use agentic paradigm for understanding and reasoning over continuous multi-view images and videos, which has the potential to improve real-world spatial intelligence applications.

📅 Published on Jun 18

🔗 Links:
• GitHub: https://github.com/huggingface
• arXiv: https://arxiv.org/abs/2606.20515
• PDF: https://arxiv.org/pdf/2606.20515
• Project Page: https://ropedia.github.io/S-Agent

━━━━━━━━━━━━━━━━━━━━━━━━
📢 By: https://xn--r1a.website/PaperNexus

#SpatialReasoning #VisualLanguageModels #3DWorldUnderstanding #SpatioTemporalEvidence #ToolUseInAI