SpaceAct is a new benchmark that tests multimodal large language models on 3D spatial and temporal reasoning through closed‑loop physical action and active perception. It uses interlocking bricks and mechanical puzzles to create 108 verified instances across 36 task families, requiring models to reconstruct 3D geometry, anticipate dynamics, and execute precise 6‑DoF poses while avoiding collisions. The accompanying SAPA framework provides diagnostic modes and active‑perception capabilities, revealing that even with a valid plan, current models struggle with 3D visual grounding and action refinement.
By Chenglong Shi, Zekai Yuan, Yue Song, Jon Pu, Will Wu
arXiv:2609.12606v2 Announce Type: replace
Abstract: While multimodal reasoning has advanced rapidly, solving complex geometry problems critically hinges on active visual assistance, such as construct...
By Zhitong Dong, Jicai Pan, Yingguo Gao, Jingting Ding, Hao Chen, Jinjie Gu
SpatialScout is a benchmark that tests vision‑language model agents on active, object‑level 3D mapping. It includes 80 indoor scenes with 3,140 target objects, where agents must control a camera, gather RGB observations, and estimate object centers in a shared coordinate frame. Performance is measured by localization accuracy and target coverage using object‑scale overlap scores, revealing that current VLMs, including Claude Opus 5.5, struggle to build precise metric maps.
By Zhaoxi Zhang, Hao Liang, Linxuan Wu, Yiming Wu, Yifu Yuan, Kun Liang, Chenming Tang, Meiyi Qiang, Wentao Zhang, Weikang Li
VABench is a benchmark that tests general‑purpose multimodal large language models (MLLMs) on embodied spatial intelligence by requiring them to observe, reason, act, and revise based on visual demonstrations and active perception. The benchmark includes 14 task families, a fixed model‑agnostic controller, and evaluates models on target localization, spatial relations, and long‑horizon composition tracks without providing privileged object poses or learned action heads. Results show that while the best model achieves perfect target localization, overall task success remains modest, and active camera control and geometric transfer significantly influence performance.
By Zhongbo Zhang, Jiayi Jin, Yifan Wang, Zaibin Zhang, Haiwen Diao, Lijun Wang, Huchuan Lu
VoxProg introduces voxel programming for 3D spatial reasoning, enabling language models to translate task interpretations into executable programs that operate on local voxel views of scene surfaces. By indexing surface geometry on regular grids, the approach isolates task-relevant regions and boundaries, allowing downstream pose programs to compute target SE(3) poses. The method is evaluated on the 71-task VoxProg-Bench benchmark, showing consistent superiority over relation-based, keypoint-based, and mesh-based baselines.
By Zhichao Zhai, Menglei Zhao, Guikun Chen, Zihao Yan, Kanzhi Wu, Wenguan Wang, Jun Xiao
arXiv:2606. 17539v1 Announce Type: cross Abstract: Spatial VLMs have made substantial progress in geometric perception, yet complex spatial reasoning requiring multi-step inference over depth, distance, and scene relations remains challenging.
By Yatai Ji, An-Chieh Cheng, Yang Fu, Yukang Chen, Han Zhang, Zhaojing Yang, Wei Huang, Ka Chun Cheung, Song Han, Vidya Nariyambut Murali, Pavlo Molchanov, Jan Kautz, Simon See, Hongxu Yin, Ping Luo, Sifei Liu
The paper introduces GeoReasoner, a framework for Aerial Vision-and-Language Navigation that replaces reactive trial-and-error with geospatial-guided active reasoning. It first uses pre-search geospatial reasoning to generate region-level priors and limit the goal hypothesis space, then performs hypothesis-driven exploration to gather discriminative evidence, and finally applies consistency-aware goal verification to avoid premature confirmations. Experiments on the CityNav benchmark show GeoReasoner achieves state‑of‑the‑art zero‑shot performance, rivaling supervised methods.
By Jiawei Ge, Wenkang Gong, Jiuxin Cao, Bo Liu, Lei Cui, Haoran Wang, Xuelin Zhu, Biwei Cao, Chang Liu, Chen Feng, Ioannis Patras
SceneTeract is a verification interface that separates semantic action understanding from physical feasibility in indoor 3D scenes. It decomposes activities into atomic actions and performs explicit geometric checks to determine executability, providing diagnostic traces for failures. The system reveals widespread functional and accessibility issues in synthetic scenes, shows that existing VLMs over‑predict action feasibility, and improves VLM performance through post‑training with verifier feedback, with benefits that generalize to real‑world scenes.
By L\'eopold Maillard, Francis Engelmann, Tom Durand, Boxiao Pan, Yang You, Leonidas Guibas, Maks Ovsjanikov
arXiv:2607. 22732v1 Announce Type: new Abstract: LLM-based game agents often perform poorly on more complex tasks.
By Mohit Jiwatode, Ronja Fuchs, Robin Schm\"ocker, Bodo Rosenhahn, Alexander Dockhorn
arXiv:2610.10390v1 Announce Type: new
Abstract: Vision-language-action~(VLA) models have emerged as a promising paradigm for autonomous driving. However, existing VLA models still suffer from a funda...
By Xingtai Gui, Yucheng Zhou, Dongqian Guo, Jiahao Gong, Feiyang Tan, Jianbing Shen
arXiv:2610.01999v1 Announce Type: new
Abstract: Spatial reasoning benchmarks evaluate vision-language models across diverse tasks, but task-level scores do not reveal which underlying capabilities ac...
By Pengzhan Sun, Junbin Xiao, Ramanathan Rajaraman, Shiu-hong Kao, Angela Yao
arXiv:2605.12413v4 Announce Type: replace
Abstract: Multimodal Large Language Models (MLLMs) show strong visual perception, yet remain limited in reasoning about space under changing viewpoints. We s...
By Yuangong Chen, Wai Keung Wong, Jiaxing Li, Ioannis Patras, Xu Zheng