arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2512.05809cs.CVcs.AI
收藏

通过测试时缩放探查世界模型在空间推理中的有效性

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)

Saurav Jha, M. Jehanzeb Mirza, Wei Lin, Shiqi Yang, Sarath Chandar

AI总结:

本文提出ViSA框架,通过可验证的微断言改进世界模型的空间推理能力,但发现当前模型在复杂任务中仍存在信息瓶颈。

摘要:

视觉-语言模型(VLMs)在需要多视角理解和具身视角转换的空间推理任务中仍然有限。最近的方法,如MindJourney,试图通过测试时缩放来弥合这一差距,其中世界模型想象动作条件轨迹,而启发式验证器从这些轨迹中选择有用的视角。在本工作中,我们系统地考察了此类测试时验证器在基准测试中的行为,揭示了它们的潜力和陷阱。我们的基于不确定性的分析显示,MindJourney的验证器提供很少有意义的校准,随机评分往往能同样有效地减少答案熵,从而暴露系统性的动作偏差和不可靠的奖励信号。为缓解这些问题,我们引入了一种通过空间断言进行验证(ViSA)框架,将测试时奖励建立在可验证、帧锚定的微断言上。这种原则性的验证器在SAT-Real基准上一致提高了空间推理能力,并通过更平衡的探索行为纠正了轨迹选择偏差。然而,在具有挑战性的MMSI-Bench上,没有任何验证器,包括我们的,都能实现一致的缩放,这表明当前的世界模型形成了一种信息瓶颈,想象的视角无法丰富细粒度推理。这些发现勾勒出测试时验证在基于世界模型推理中的好坏和丑陋方面。我们的代码可在https://github.com/chandar-lab/visa-for-mindjourney获取。

英文摘要:

Vision-Language Models (VLMs) remain limited in spatial reasoning tasks that require multi-view understanding and embodied perspective shifts. Recent approaches such as MindJourney attempt to mitigate this gap through test-time scaling where a world model imagines action-conditioned trajectories and a heuristic verifier selects helpful views from such trajectories. In this work, we systematically examine how such test-time verifiers behave across benchmarks, uncovering both their promise and their pitfalls. Our uncertainty-based analyses show that MindJourney's verifier provides little meaningful calibration, and that random scoring often reduces answer entropy equally well, thus exposing systematic action biases and unreliable reward signals. To mitigate these, we introduce a Verification through Spatial Assertions (ViSA) framework that grounds the test-time reward in verifiable, frame-anchored micro-claims. This principled verifier consistently improves spatial reasoning on the SAT-Real benchmark and corrects trajectory-selection biases through more balanced exploratory behavior. However, on the challenging MMSI-Bench, none of the verifiers, including ours, achieve consistent scaling, suggesting that the current world models form an information bottleneck where imagined views fail to enrich fine-grained reasoning. Together, these findings chart the bad, good, and ugly aspects of test-time verification for world-model-based reasoning. Our code is available at https://github.com/chandar-lab/visa-for-mindjourney.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑