Despite rapid progress in MLLMs, visual spatial reasoning remains unreliable when correct answers depend on how a scene would appear under unseen or alternative viewpoints. Recent work addresses this by augmenting reasoning with world models for visual imagination, but questions such as when imagination is actually necessary, how much of it is beneficial, and when it becomes harmful, remain poorly understood. In practice, indiscriminate imagination can increase computation and even degrade performance by introducing misleading evidence. In this work, we present an in-depth analysis of test-time visual imagination as a controllable resource for spatial reasoning. We first study when static visual evidence is sufficient, when imagination improves reasoning, and how excessive or unnecessary imagination affects accuracy and efficiency. To support this analysis, we then introduce AVIC, an adaptive test-time framework with world models that explicitly reasons about the sufficiency of current visual evidence before selectively invoking and scaling visual imagination. Finally, to further learn this gating and planning behavior without any annotation of when and how much to imagine, we introduce AVIC-R, which trains the policy via GRPO from QA-correctness rewards and penalties by imagination cost. Across spatial reasoning benchmarks (SAT, MMSI) and an embodied navigation benchmark (R2R), our results reveal clear scenarios where imagination is critical, marginal, or detrimental, and show that selective control can match or outperform fixed imagination strategies with substantially fewer world-model calls and language tokens. Our AVIC-R surpasses strong proprietary baselines including GPT-4o and GPT-4.1 while invoking the world model less often. Overall, our findings highlight the importance of analyzing and controlling test-time imagination for efficient and reliable spatial reasoning.
Imagine Before You Draw: Visual Prompt Engineering for Image Generation
Nanyang Technological University（南洋理工大学）；National University of Singapore（国立新加坡大学）；Zhejiang University（浙江大学）；The Chinese University of Hong Kong（香港中文大学）
From Imagined Futures to Executable Actions: Mixture of Latent Actions for Robot Manipulation
School of Data Science, Fudan University（复旦大学数据科学学院）；Shanghai Innovation Institute（上海创新研究院）；Imperial College London（伦敦帝国理工学院）；University of Surrey（萨里大学）
Probing the effectiveness of World Models for Spatial Reasoning through Test-time Scaling
MILA – Quebec AI Institute（魁北克人工智能研究所）；Polytechnique Montréal（蒙特利尔理工学院）；MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）；Institute for Machine Learning, Johannes Kepler University Linz（林茨约瑟夫·夫兰克大学机器学习研究所）；Nankai University（南开大学）
Mixture-of-Visual-Thoughts: Exploring Context-Adaptive Reasoning Mode Selection for General Visual Reasoning
Fudan University（复旦大学）；Alibaba Group Holding Limited（阿里巴巴集团控股有限公司）；Future Living Lab of Alibaba（阿里巴巴未来生活实验室）；University of Southern California（南加州大学）；Shanghai Innovation Institute（上海创新研究院）
Zero-Shot Visual Reasoning by Vision-Language Models: Benchmarking and Analysis
ASUS Intelligent Cloud Services (AICS)（华硕智能云服务（AICS））；ASUS Global Pte. Ltd.（华硕全球私人有限公司）；Centre for Frontier AI Research (CFAR)（前沿人工智能研究中心（CFAR））；Institute of High Performance Computing (IHPC)（高性能计算研究所（IHPC））；Agency for Science, Technology and Research (A*STAR)（新加坡科学、技术与研究局（A*STAR））
Understanding the computational demands underlying visual reasoning
Artificial and Natural Intelligence Toulouse Institute（图卢兹人工智能与自然智能研究所）；Université de Toulouse（图卢兹大学）；Centre de Recherche Cerveau & Cognition (CerCo), CNRS（法国国家科学研究中心大脑与认知研究中心）
MedLVR: Latent Visual Reasoning for Reliable Medical Visual Question Answering
Department of Radiation Oncology and Winship Cancer Institute, Emory University School of Medicine（埃默里大学医学院放射肿瘤学系与温希普癌症研究所）；Department of Biostatistics and Bioinformatics, Emory University（埃默里大学生物统计学与生物信息学系）
ReasonCLIP-58M: Visually Grounded Commonsense Reasoning Supervision for CLIP
Khalifa University（卡利法大学）；University of Western Australia（西澳大学）；The Chinese University of Hong Kong（香港中文大学）；University of Melbourne（墨尔本大学）；University of Central Florida（佛罗里达中央大学）
Look Again, Think Slowly: Enhancing Visual Reflection in Vision-Language Models
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Wuhan AI Research（武汉人工智能研究院）
Mitigating Visual Forgetting via Take-along Visual Conditioning for Multi-modal Long CoT Reasoning
School of Artificial Intelligence, Nanjing University（南京大学人工智能学院）；National Key Laboratory for Novel Software Technology, Nanjing University（南京大学新型软件技术国家重点实验室）；Tencent（腾讯）；Center for Language AI Research, Tohoku University（东北大学语言人工智能研究中心）；RIKEN Center for Advanced Intelligence Project（日本理化学研究所先进人工智能项目中心）
Imagine by Reasoning: A Reasoning-Based Implicit Semantic Data Augmentation for Long-Tailed Classification
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）
GAMR: A Guided Attention Model for (visual) Reasoning
Carney Institute for Brain Science（卡尼脑科学研究所）；Brown University（布朗大学）；Centre de Recherche Cerveau et Cognition（脑与认知研究中心）；CNRS（法国国家科学研究中心）；Université de Toulouse（图卢兹大学）；Artificial and Natural Intelligence Toulouse Institute（图卢兹人工智能与自然智能研究所）
Thinking with Imagination: Agentic Visual Spatial Reasoning with World Simulators
The University of Hong Kong（香港大学）；Shanghai AI Laboratory（上海人工智能实验室）；Shanghai Jiao Tong University（上海交通大学）；Fudan University（复旦大学）；Beihang University（北航大学）
