SPARGen: Unifying Spatial Perception and Reasoning through Native Multimodal Generation
Summary
SPARGen introduces a unified multimodal framework that casts 3D reconstruction, dense correspondence, and spatial reasoning as instruction-conditioned generation tasks, enabling shared spatial representations.
View Cached Full Text
Cached at: 08/17/26, 03:45 AM
Paper page - SPARGen: Unifying Spatial Perception and Reasoning through Native Multimodal Generation
Source: https://huggingface.co/papers/2608.14138
Abstract
SPARGen unifies 3D reconstruction, dense correspondence, and spatial reasoning into a single instruction-conditioned multimodal generative model that jointly learns shared spatial representations.
Spatial perception and reasoning from visual observations require recovering geometric structure, establishing correspondences, and understanding spatial relations. Existing approaches typically address these capabilities separately using task-specific architectures or external geometric modules, limiting knowledge transfer among complementary representations of the same physical scene. We introduce SPARGen, a unifiedmultimodal frameworkthat casts3D reconstruction,dense correspondence, andspatial reasoningasinstruction-conditioned generationtasks. SPARGen serializes compact structured and linguistic outputs astoken sequenceswhile generating densegeometric fieldsin image-aligned forms, enablingspatial supervisionto jointly shape shared representations within anative multimodal generative model. Experiments across benchmarks for3D reconstruction, correspondence, andspatial reasoningshow that SPARGen achieves competitive performance across heterogeneous spatial tasks within a single native multimodal generative framework.
View arXiv pageView PDFAdd to collection
Get this paper in your agent:
hf papers read 2608\.14138
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2608.14138 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2608.14138 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2608.14138 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Show, Don't Tell: Evaluating Spatial Cognition in Generative Pixels Rather Than LLM Text
Proposes ProVisE, a benchmark-agnostic framework to evaluate spatial cognition in image-generation models using pixel-space outputs, and introduces SpatialGen-Bench for unified evaluation across 14 spatial subtasks.
Reinforcing Dual-Path Reasoning in Spatial Vision Language Models
This paper introduces SR-REAL, a unified framework for spatial vision-language models that combines linguistic deduction and 3D geometric reasoning via reinforcement learning, enabling robust multi-step spatial reasoning across diverse tasks.
Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoning
A training-free framework for spatial reasoning from egocentric videos that enables revisiting conclusions through synthesized novel-view videos generated from predicted 3D geometry.
SpaR3D-MoE: Adaptive 3D Spatial Reasoning from Sparse Views Meets Geometry-Inductive Mixture-of-Experts
Introduces SpaR3D-MoE, an end-to-end framework for adaptive 3D spatial reasoning from sparse RGB views, using manifold sampling and geometry-inductive mixture-of-experts to achieve state-of-the-art performance on VSI-Bench, ScanQA, and SQA3D.
Disentangling 3D Modeling from Spatial Reasoning
This paper proposes DiSR, a framework that separates 3D perception from reasoning by using off-the-shelf perception models to reconstruct explicit 3D evidence and fine-tuning an LLM with LoRA for spatial reasoning, achieving competitive performance with improved interpretability and efficiency.