SpatialThinker: Reinforcing 3D Reasoning in Multimodal LLMs via Spatial Rewards

• URL: http://arxiv.org/abs/2511.07403v1
• Date: Mon, 10 Nov 2025 18:52:47 GMT
• Title: SpatialThinker: Reinforcing 3D Reasoning in Multimodal LLMs via Spatial Rewards
• Authors: Hunar Batra, Haoqin Tu, Hardy Chen, Yuanze Lin, Cihang Xie, Ronald Clark,
• Abstract summary: We introduce SpatialThinker, a 3D-aware MLLM trained with RL to integrate structured spatial grounding with multi-step reasoning.<n>The model simulates human-like spatial perception by constructing a scene graph of task-relevant objects and spatial relations, and reasoning towards an answer via dense spatial rewards.
• Score: 37.39035418889281
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal large language models (MLLMs) have achieved remarkable progress in vision-language tasks, but they continue to struggle with spatial understanding. Existing spatial MLLMs often rely on explicit 3D inputs or architecture-specific modifications, and remain constrained by large-scale datasets or sparse supervision. To address these limitations, we introduce SpatialThinker, a 3D-aware MLLM trained with RL to integrate structured spatial grounding with multi-step reasoning. The model simulates human-like spatial perception by constructing a scene graph of task-relevant objects and spatial relations, and reasoning towards an answer via dense spatial rewards. SpatialThinker consists of two key contributions: (1) a data synthesis pipeline that generates STVQA-7K, a high-quality spatial VQA dataset, and (2) online RL with a multi-objective dense spatial reward enforcing spatial grounding. SpatialThinker-7B outperforms supervised fine-tuning and the sparse RL baseline on spatial understanding and real-world VQA benchmarks, nearly doubling the base-model gain compared to sparse RL, and surpassing GPT-4o. These results showcase the effectiveness of combining spatial supervision with reward-aligned reasoning in enabling robust 3D spatial understanding with limited data and advancing MLLMs towards human-level visual reasoning.

Related papers

• SpatialMosaic: A Multiview VLM Dataset for Partial Visibility [25.874299974251965]
  We propose a scalable multi-view data generation and annotation pipeline that constructs realistic spatial reasoning QAs.<n>We introduce SpatialMosaic-Bench, a benchmark for evaluating multi-view spatial reasoning under realistic and challenging scenarios.<n>We also present SpatialMosaicVLM, a hybrid framework that integrates 3D reconstruction models as geometry encoders within Vision-Language Models.
  arXiv  Detail & Related papers  (2025-12-29T10:48:54Z)
• SpatialGeo:Boosting Spatial Reasoning in Multimodal LLMs via Geometry-Semantics Fusion [23.86761713752287]
  Multimodal large language models (MLLMs) have achieved significant progress in image and language tasks.<n>Most MLLMs suffer from limited spatial reasoning ability to interpret and infer spatial arrangements in three-dimensional space.<n>We propose a novel vision encoder based on hierarchical fusion of geometry and semantics features, generating spatial-aware visual embedding.
  arXiv  Detail & Related papers  (2025-11-21T15:24:33Z)
• Actial: Activate Spatial Reasoning Ability of Multimodal Large Language Models [75.45940282834327]
  We introduce Viewpoint Learning, a task designed to evaluate and improve the spatial reasoning capabilities of MLLMs.<n>We present the Viewpoint-100K dataset, consisting of 100K object-centric image pairs with diverse viewpoints and corresponding question-answer pairs.<n>Our approach employs a two-stage fine-tuning strategy, resulting in significant improvements across multiple tasks.
  arXiv  Detail & Related papers  (2025-11-03T14:27:00Z)
• Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence [13.168559963356952]
  We present Spatial-MLLM, a novel framework for visual-based spatial reasoning from purely 2D observations.<n>Our key insight is to unleash the strong structure prior to the feed-forward visual geometry foundation model.<n>A connector then integrates both features into unified visual tokens for enhanced spatial understanding.
  arXiv  Detail & Related papers  (2025-05-29T17:59:04Z)
• ViewSpatial-Bench: Evaluating Multi-perspective Spatial Localization in Vision-Language Models [68.46716645478661]
  Vision-language models (VLMs) have demonstrated remarkable capabilities in understanding and reasoning about visual content.<n>Current VLMs excel primarily at egocentric spatial reasoning (from the camera's perspective) but fail to generalize to allocentric viewpoints.<n>We introduce ViewSpatial-Bench, the first comprehensive benchmark designed specifically for multi-viewpoint spatial localization recognition evaluation.
  arXiv  Detail & Related papers  (2025-05-27T17:59:26Z)
• SpatialScore: Towards Unified Evaluation for Multimodal Spatial Understanding [64.15606979785355]
  Multimodal large language models (MLLMs) have achieved impressive success in question-answering tasks, yet their capabilities for spatial understanding are less explored.<n>This work investigates a critical question: do existing MLLMs possess 3D spatial perception and understanding abilities?
  arXiv  Detail & Related papers  (2025-05-22T17:59:03Z)
• SpaceR: Reinforcing MLLMs in Video Spatial Reasoning [70.7401015322983]
  Video spatial reasoning poses a significant challenge for existing Multimodal Large Language Models (MLLMs)<n>This limitation stems primarily from 1) the absence of high-quality datasets for this task, and 2) the lack of effective training strategies to develop spatial reasoning capabilities.<n>Motivated by the success of Reinforcement Learning with Verifiable Reward (RLVR) in unlocking spatial reasoning abilities, this aims to improve MLLMs in video spatial reasoning through the RLVR paradigm.
  arXiv  Detail & Related papers  (2025-04-02T15:12:17Z)
• From Flatland to Space: Teaching Vision-Language Models to Perceive and Reason in 3D [32.547597353581594]
  We introduce a novel 2D spatial data generation and annotation pipeline built upon scene data with 3D ground-truth.<n>We construct SPAR-7M, a large-scale dataset generated from thousands of scenes across multiple public datasets.<n>In addition, we introduce SPAR-Bench, a benchmark designed to offer a more comprehensive evaluation of spatial capabilities.
  arXiv  Detail & Related papers  (2025-03-29T04:51:50Z)
• Open3D-VQA: A Benchmark for Comprehensive Spatial Reasoning with Multimodal Large Language Model in Open Space [38.482463743451625]
  We present Open3D-VQA, a novel benchmark for evaluating MLLMs' ability to reason about complex spatial relationships from an aerial perspective.<n>The benchmark comprises 73k QA pairs spanning 7 general spatial reasoning tasks, including multiple-choice, true/false, and short-answer formats.
  arXiv  Detail & Related papers  (2025-03-14T05:35:38Z)
• Sparkle: Mastering Basic Spatial Capabilities in Vision Language Models Elicits Generalization to Spatial Reasoning [36.588008658084895]
  Vision language models (VLMs) perform well on many tasks but often fail at spatial reasoning.<n>Our evaluation shows that state-of-the-art VLMs give implausible or incorrect answers to composite spatial problems.<n>We enhance 2D spatial reasoning in VLMs by training them only on basic spatial capabilities.
  arXiv  Detail & Related papers  (2024-10-21T16:26:09Z)
• REVISION: Rendering Tools Enable Spatial Fidelity in Vision-Language Models [67.55362046790512]
  Vision-language models lack the ability to correctly reason over spatial relationships. We develop the REVISION framework which improves spatial fidelity in vision-language models. Our results and findings indicate that utilizing rendering-based frameworks is an effective approach for developing spatially-aware models.
  arXiv  Detail & Related papers  (2024-08-05T04:51:46Z)
• SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities [59.39858959066982]
  understanding and reasoning about spatial relationships is a fundamental capability for Visual Question Answering (VQA) and robotics. We develop an automatic 3D spatial VQA data generation framework that scales up to 2 billion VQA examples on 10 million real-world images. By training a VLM on such data, we significantly enhance its ability on both qualitative and quantitative spatial VQA.
  arXiv  Detail & Related papers  (2024-01-22T18:01:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.