Thinking with Novel Views: A Systematic Analysis of Generative-Augmented Spatial Intelligence

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Yanbing, Wang, Bo, Liu, Jianhui, Jiang, Nan, Jiang, Jiaxiu, Sun, Haoze, Yang, Yijun, Zheng, Shenghe, Song, Lin, Huang, Haoyang, Duan, Nan, Li, Wenbo
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917480664924160
author Zhang, Yanbing
Wang, Bo
Liu, Jianhui
Jiang, Nan
Jiang, Jiaxiu
Sun, Haoze
Yang, Yijun
Zheng, Shenghe
Song, Lin
Huang, Haoyang
Duan, Nan
Li, Wenbo
author_facet Zhang, Yanbing
Wang, Bo
Liu, Jianhui
Jiang, Nan
Jiang, Jiaxiu
Sun, Haoze
Yang, Yijun
Zheng, Shenghe
Song, Lin
Huang, Haoyang
Duan, Nan
Li, Wenbo
contents Current Large Multimodal Models (LMMs) struggle with spatial reasoning tasks requiring viewpoint-dependent understanding, largely because they are confined to a single, static observation. We propose Thinking with Novel Views (TwNV), a paradigm that integrates generative novel-view synthesis into the reasoning loop: a Reasoner LMM identifies spatial ambiguity, instructs a Painter to synthesize an alternative viewpoint, and re-examines the scene with the additional evidence. Through systematic experiments we address three research questions. (1) Instruction format: numerical camera-pose specifications yield more reliable view control than free-form language. (2) Generation fidelity: synthesized view quality is tightly coupled with downstream spatial accuracy. (3) Inference-time visual scaling: iterative multi-turn view refinement further improves performance, echoing recent scaling trends in language reasoning. Across four spatial subtask categories and four LMM architectures (both closed- and open-source), TwNV consistently improves accuracy by +1.3 to +3.9 pp, with the largest gains on viewpoint-sensitive subtasks. These results establish novel-view generation as a practical lever for advancing spatial intelligence of LMMs.
format Preprint
id arxiv_https___arxiv_org_abs_2605_10588
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Thinking with Novel Views: A Systematic Analysis of Generative-Augmented Spatial Intelligence
Zhang, Yanbing
Wang, Bo
Liu, Jianhui
Jiang, Nan
Jiang, Jiaxiu
Sun, Haoze
Yang, Yijun
Zheng, Shenghe
Song, Lin
Huang, Haoyang
Duan, Nan
Li, Wenbo
Computer Vision and Pattern Recognition
Current Large Multimodal Models (LMMs) struggle with spatial reasoning tasks requiring viewpoint-dependent understanding, largely because they are confined to a single, static observation. We propose Thinking with Novel Views (TwNV), a paradigm that integrates generative novel-view synthesis into the reasoning loop: a Reasoner LMM identifies spatial ambiguity, instructs a Painter to synthesize an alternative viewpoint, and re-examines the scene with the additional evidence. Through systematic experiments we address three research questions. (1) Instruction format: numerical camera-pose specifications yield more reliable view control than free-form language. (2) Generation fidelity: synthesized view quality is tightly coupled with downstream spatial accuracy. (3) Inference-time visual scaling: iterative multi-turn view refinement further improves performance, echoing recent scaling trends in language reasoning. Across four spatial subtask categories and four LMM architectures (both closed- and open-source), TwNV consistently improves accuracy by +1.3 to +3.9 pp, with the largest gains on viewpoint-sensitive subtasks. These results establish novel-view generation as a practical lever for advancing spatial intelligence of LMMs.
title Thinking with Novel Views: A Systematic Analysis of Generative-Augmented Spatial Intelligence
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.10588