Newest papers

MindView: Benchmarking Generative Multi-View Spatial Intelligence

Yanbo Wang, Siqi Luo, Siyu Zhang, Chenming Zhu, Wenzhao Zheng, Qian Yu, Juncheng Yan, Bin Fu, Tai Wang, Yu Qiao, Yihao Liu, Jiwen Lu · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28622 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

O benefício da inteligência de contextos espaciais selecionados foi recentemente avaliado por meio de tarefas gerativas como a edição de imagens espacialmente fundamentadas, mas a geração de visões espacialmente consistentes por meio de observações de múltiplas visões permanece inexplorada. Introduzimos o MindView, um benchmark para a inteligência espacial multi-visão gerativa com 1.500 amostras de 513 cenas em três conjuntos de dados internos. Suas sete tarefas formam duas famílias: tarefas geométricas-dominantes especificam mudanças na posição e na orientação da câmara, enquanto as tarefas semânticas-dominantes exigem inferir pontos de visão alvo por meio da aterrissagem de objetos espaciais referenciados e da interpretação de relações espaciais e dicas de referência

Machine translation (Meta M2M100); the English original follows.

Abstract

Spatial intelligence has recently been assessed through generative tasks such as spatially grounded image editing, yet language-guided generation of spatially consistent views from partial multi-view observations remains underexplored. We introduce MindView, a benchmark for generative multi-view spatial intelligence with 1,500 samples from 513 scenes across three indoor datasets. Its seven tasks form two families: geometry-dominant tasks specify changes in camera position and orientation, while semantics-dominant tasks require inferring target viewpoints by grounding referred objects and interpreting spatial relations and reference-frame cues. Evaluation of eleven generators reveals substantial room for improvement on MindView. A paired context analysis further shows that all eleven generators obtain lower overall scores under the expanded context setting. These findings motivate us to investigate whether task-specific supervision can improve target-view generation and whether models can use additional multi-view evidence effectively. We therefore fine-tune BAGEL on the MindView training dataset, improving its overall score by 24.3 points and yielding gains across all seven MindView tasks. Beyond these generation gains, the fine-tuned model also yields higher scores on three spatial-understanding benchmarks without training on their annotations, suggesting that learning to generate spatially grounded views may benefit spatial understanding. To improve evidence utilization under expanded context, we propose MindCue, a training-free spatial evidence selector guided by a target-view hypothesis. MindCue reverses the drop in overall score under expanded context and outperforms the standard-context baseline. Overall, MindView provides a concrete testbed for generative multi-view spatial intelligence.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Spatial contextual awareness, Artificial intelligence, Generative grammar, Benchmarking, Machine learning

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange