Variable-length autoregressive latent sequences, trained as variational inference with a PPO-style objective, give robot policies adaptive test-time compute and yield a reusable action tokenizer.
hub
SpatialVLM: Endowing vision-language models with spatial reasoning capabilities
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
abstract
Understanding and reasoning about spatial relationships is a fundamental capability for Visual Question Answering (VQA) and robotics. While Vision Language Models (VLM) have demonstrated remarkable performance in certain VQA benchmarks, they still lack capabilities in 3D spatial reasoning, such as recognizing quantitative relationships of physical objects like distances or size differences. We hypothesize that VLMs' limited spatial reasoning capability is due to the lack of 3D spatial knowledge in training data and aim to solve this problem by training VLMs with Internet-scale spatial reasoning data. To this end, we present a system to facilitate this approach. We first develop an automatic 3D spatial VQA data generation framework that scales up to 2 billion VQA examples on 10 million real-world images. We then investigate various factors in the training recipe, including data quality, training pipeline, and VLM architecture. Our work features the first internet-scale 3D spatial reasoning dataset in metric space. By training a VLM on such data, we significantly enhance its ability on both qualitative and quantitative spatial VQA. Finally, we demonstrate that this VLM unlocks novel downstream applications in chain-of-thought spatial reasoning and robotics due to its quantitative estimation capability. Project website: https://spatial-vlm.github.io/
hub tools
citation-role summary
background 2
citation-polarity summary
years
2026 15roles
background 2polarities
background 2representative citing papers
Authors create ReasonMatch-Bench and DCRL training to boost MLLM performance on wide-baseline matching, reporting gains over baselines while preserving general capabilities.
World models succeed when their latent states are built to meet task-specific sufficiency constraints rather than preserving the maximum amount of information.
Fine-tuning multimodal models on a new synthetic spatial benchmark improves generative spatial compliance on real and synthetic tasks and transfers to better spatial understanding.
Agentic RAG-VLM achieves 78.3% success on a 12-task grasping benchmark with 360 trials per configuration, a 53.3 percentage-point gain over VLM-only baselines, via hierarchical affordance RAG, scene graph constraints, and a 14-type failure taxonomy with adaptive retry.
SR-REAL equips spatial VLMs with dual LOR and DTR reasoning paths trained via RL, achieving better benchmark performance through mutual reinforcement and generalization without per-task tuning.
Reasmory turns 3D reconstruction into validated program-executable memory for VLMs, yielding 6-18% gains on spatial reasoning benchmarks over direct baselines.
PanoWorld adds spherical spatial cross-attention and pano-native training data to MLLMs for improved spatial reasoning on ERP panoramas, outperforming baselines on new and existing benchmarks.
Multimodal LLMs significantly underperform humans at spotting objects that break 3D consistency in multi-view image pairs.
TrianguLang achieves state-of-the-art feed-forward text-guided 3D localization and segmentation by using predicted geometry to gate cross-view semantic correspondences without ground-truth poses.
HoloAgent-0 is a unified embodied agent framework with Embodied AgentOS, 3D spatial memory, and embodied skills, deployed and evaluated on real robot hardware for navigation and manipulation tasks.
GeoVR distills camera pose, depth, scale, and multi-scale 3D features from pre-trained models into MLLMs via video supervision to improve spatial reasoning.
Presents LongSpace-Bench benchmark and LongSpace framework that chunks long videos, adds 3D structural cues, and builds layer-aware memory to improve spatial reasoning in multimodal LLMs.
GeoAlign post-trains an RGB geometry branch on robot RGB-D data to produce GEP features that are queried by proprioceptive state to generate phase-dependent geometry tokens, yielding 99.0% on LIBERO, 85.3% on SimplerEnv-Fractal, and 78.8% on real ALOHA tasks.
Core identification in TTC markets is solvable in O(Ln) time via randomized SVD on a Markov transition matrix, asymptotically optimal for sparse preferences and inheriting TTC properties.
citing papers explorer
-
Latent Memory Palace: Reasoning for Control as Autoregressive Variational Inference
Variable-length autoregressive latent sequences, trained as variational inference with a PPO-style objective, give robot policies adaptive test-time compute and yield a reusable action tokenizer.
-
Eliciting Complex Spatial Reasoning in MLLMs through Wide-Baseline Matching
Authors create ReasonMatch-Bench and DCRL training to boost MLLM performance on wide-baseline matching, reporting gains over baselines while preserving general capabilities.
-
Latent State Design for World Models under Sufficiency Constraints
World models succeed when their latent states are built to meet task-specific sufficiency constraints rather than preserving the maximum amount of information.
-
Exploring Spatial Intelligence from a Generative Perspective
Fine-tuning multimodal models on a new synthetic spatial benchmark improves generative spatial compliance on real and synthetic tasks and transfers to better spatial understanding.
-
Agentic RAG-VLM: Affordance-Aware Retrieval-Augmented Generation with Self-Reflective Planning for Robotic Grasping
Agentic RAG-VLM achieves 78.3% success on a 12-task grasping benchmark with 360 trials per configuration, a 53.3 percentage-point gain over VLM-only baselines, via hierarchical affordance RAG, scene graph constraints, and a 14-type failure taxonomy with adaptive retry.
-
Reinforcing Dual-Path Reasoning in Spatial Vision Language Models
SR-REAL equips spatial VLMs with dual LOR and DTR reasoning paths trained via RL, achieving better benchmark performance through mutual reinforcement and generalization without per-task tuning.
-
Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning
Reasmory turns 3D reconstruction into validated program-executable memory for VLMs, yielding 6-18% gains on spatial reasoning benchmarks over direct baselines.
-
PanoWorld: Towards Spatial Supersensing in 360$^\circ$ Panorama World
PanoWorld adds spherical spatial cross-attention and pano-native training data to MLLMs for improved spatial reasoning on ERP panoramas, outperforming baselines on new and existing benchmarks.
-
Multimodal Language Models Cannot Spot Spatial Inconsistencies
Multimodal LLMs significantly underperform humans at spotting objects that break 3D consistency in multi-view image pairs.
-
TrianguLang: Geometry-Aware Semantic Consensus for Pose-Free 3D Localization
TrianguLang achieves state-of-the-art feed-forward text-guided 3D localization and segmentation by using predicted geometry to gate cross-view semantic correspondences without ground-truth poses.
-
HoloAgent-0: A Unified Embodied Agent Framework with 3D Spatial Memory
HoloAgent-0 is a unified embodied agent framework with Embodied AgentOS, 3D spatial memory, and embodied skills, deployed and evaluated on real robot hardware for navigation and manipulation tasks.
-
Learning Geometric Representations from Videos for Spatial Intelligent Multimodal Large Language Models
GeoVR distills camera pose, depth, scale, and multi-scale 3D features from pre-trained models into MLLMs via video supervision to improve spatial reasoning.
-
LongSpace: Exploring Long-Horizon Spatial Memory from Perception to Recall in Video
Presents LongSpace-Bench benchmark and LongSpace framework that chunks long videos, adds 3D structural cues, and builds layer-aware memory to improve spatial reasoning in multimodal LLMs.
-
GeoAlign: Beyond Semantics with State-Guided Spatial Alignment in VLA Models
GeoAlign post-trains an RGB geometry branch on robot RGB-D data to produce GEP features that are queried by proprioceptive state to generate phase-dependent geometry tokens, yielding 99.0% on LIBERO, 85.3% on SimplerEnv-Fractal, and 78.8% on real ALOHA tasks.
-
Fast Core Identification
Core identification in TTC markets is solvable in O(Ln) time via randomized SVD on a Markov transition matrix, asymptotically optimal for sparse preferences and inheriting TTC properties.