DeepLatent introduces a parallel latent visual reasoning framework with learnable 2D tokens and continuous RL, trained via distillation then RL, plus a new 180K dataset, claiming SOTA benchmark results.
hub Canonical reference
Think with 3d: Geometric imagination grounded spatial reasoning from limited views.arXiv preprint arXiv:2510.18632, 2025a
Canonical reference. 100% of citing Pith papers cite this work as background.
22
Pith papers citing it
Background
100% of classified citations
hub tools
citation-role summary
background 7
citation-polarity summary
roles
background 7polarities
background 7representative citing papers
View Dropout forces reliance on intermediate thinking images in unified multimodal models, with panoramic renderings proving most effective for out-of-domain cross-view spatial reasoning.
GAMSI is a dual-pathway Geometry-Aware MLLM using Metric-Structure Decoupled Queries and Expert-Guided Visual Grounding on RGB inputs alone, trained on a new 152k-sample MTS dataset to reach SOTA on seven spatial benchmarks.
4DThinker enables VLMs to perform dynamic spatial reasoning by thinking with 4D latent mental imagery using new fine-tuning and reinforcement learning methods.
A training-free Visual Chain-of-Thought framework reconstructs high-fidelity 3D meshes from single images and iteratively synthesizes optimal novel views to enhance MLLM spatial comprehension on benchmarks like 3DSRBench.
Backward token warping in ViT-based MLLMs enables reliable reasoning from nearby viewpoints by preserving semantic coherence better than pixel-wise warping or fine-tuning baselines.
GeoThinker enables active, task-conditioned geometry integration in MLLMs via spatial-grounded fusion and importance gating, reaching 72.6 on VSI-Bench.
S²GR adds stepwise thinking tokens with contrastive supervision on codebook clusters to balance computational focus and ground reasoning paths in generative recommendation.
OmniSpace is a plug-and-play method that improves spatial reasoning in MLLMs for AV by injecting camera pose, using epipolar attention across views, and distilling 3D geometric knowledge to overcome weak cross-view correspondence and depth estimation.
A self-evolving training loop that generates its own spatial QA data with executable code and difficulty feedback lifts Qwen3-VL-4B/8B to 62.7/63.3 on VSI-Bench using an order of magnitude less data.
Reasmory turns 3D reconstruction into validated program-executable memory for VLMs, yielding 6-18% gains on spatial reasoning benchmarks over direct baselines.
GASP injects geometric priors into VLMs via a deep-supervised correspondence head trained on video point correspondences and depth consistency, raising internal matching accuracy and delivering gains on spatial benchmarks without any 3D VQA data.
ProSR adds a Counterfactual Invariance Penalty and a Tail Drift Penalty to shape VLM reasoning trajectories for better visual dependence and stability on spatial tasks.
Mull-Tokens are modality-agnostic latent tokens that enable free-form multimodal thinking and deliver up to 16% gains on spatial reasoning benchmarks.
OmniView-Space framework with MPSM, tool-guided reasoning, and distillation achieves SOTA on spatial reasoning benchmarks for MLLMs while reducing external geometry dependencies.
SpatialSV trains MLLMs to lift 2D visual features into explicit 3D representations via task-oriented supervision for interpretable spatial awareness.
Integrating generative novel-view synthesis into LMM reasoning loops improves accuracy on spatial subtasks by 1.3 to 3.9 percentage points across multiple models and tasks.
MAG-3D is a training-free multi-agent framework that coordinates planning, grounding, and coding agents with off-the-shelf VLMs to achieve grounded 3D reasoning and state-of-the-art benchmark results.
OpenSpatial supplies a principled open-source data engine and 3-million-sample dataset that raises spatial-reasoning model performance by an average of 19 percent on benchmarks.
EgoMind uses Role-Play Caption and Progressive Spatial Analysis to give MLLMs competitive multi-frame spatial reasoning without 3D priors, using only 5K SFT and 20K RL samples.
OpenWorldLib defines world models as perception-centered systems with interaction and long-term memory, and provides a modular inference codebase unifying interactive video, 3D, reasoning, and VLA tasks.
Reinforcement learning is advanced for communication-efficient federated optimization and for preference-aligned, contextually safe policies in large language models.
citing papers explorer
-
DeepLatent: Think with Images via Parallel Latent Visual Reasoning
DeepLatent introduces a parallel latent visual reasoning framework with learnable 2D tokens and continuous RL, trained via distillation then RL, plus a new 180K dataset, claiming SOTA benchmark results.
-
How and What to Imagine? Visual Thinking in Unified Multimodal Models for Cross-View Spatial Reasoning
View Dropout forces reliance on intermediate thinking images in unified multimodal models, with panoramic renderings proving most effective for out-of-domain cross-view spatial reasoning.
-
Dual-Pathway Geometry-Aware MLLM for Spatial Intelligence
GAMSI is a dual-pathway Geometry-Aware MLLM using Metric-Structure Decoupled Queries and Expert-Guided Visual Grounding on RGB inputs alone, trained on a new 152k-sample MTS dataset to reach SOTA on seven spatial benchmarks.
-
4DThinker: Thinking with 4D Imagery for Dynamic Spatial Understanding
4DThinker enables VLMs to perform dynamic spatial reasoning by thinking with 4D latent mental imagery using new fine-tuning and reinforcement learning methods.
-
Enhancing MLLM Spatial Understanding via Active 3D Scene Exploration for Multi-Perspective Reasoning
A training-free Visual Chain-of-Thought framework reconstructs high-fidelity 3D meshes from single images and iteratively synthesizes optimal novel views to enhance MLLM spatial comprehension on benchmarks like 3DSRBench.
-
Token Warping Helps MLLMs Look from Nearby Viewpoints
Backward token warping in ViT-based MLLMs enables reliable reasoning from nearby viewpoints by preserving semantic coherence better than pixel-wise warping or fine-tuning baselines.
-
Thinking with Geometry: Active Geometry Integration for Spatial Reasoning
GeoThinker enables active, task-conditioned geometry integration in MLLMs via spatial-grounded fusion and importance gating, reaching 72.6 on VSI-Bench.
-
S$^2$GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendation
S²GR adds stepwise thinking tokens with contrastive supervision on codebook clusters to balance computational focus and ground reasoning paths in generative recommendation.
-
OmniSpace: Efficient Geometry Awareness for Autonomous Vehicles MLLMs
OmniSpace is a plug-and-play method that improves spatial reasoning in MLLMs for AV by injecting camera pose, using epipolar attention across views, and distilling 3D geometric knowledge to overcome weak cross-view correspondence and depth estimation.
-
Ouroboros-Spatial: Closing the Data-Model Loop for Spatial Reasoning
A self-evolving training loop that generates its own spatial QA data with executable code and difficulty feedback lifts Qwen3-VL-4B/8B to 62.7/63.3 on VSI-Bench using an order of magnitude less data.
-
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.
-
Beyond 3D VQAs: Injecting 3D Spatial Priors into Vision-Language Models for Enhanced Geometric Reasoning
GASP injects geometric priors into VLMs via a deep-supervised correspondence head trained on video point correspondences and depth consistency, raising internal matching accuracy and delivering gains on spatial benchmarks without any 3D VQA data.
-
ProSR: Process-Shaped Spatial Reasoning for Reliable Chain-of-Thought in VLMs
ProSR adds a Counterfactual Invariance Penalty and a Tail Drift Penalty to shape VLM reasoning trajectories for better visual dependence and stability on spatial tasks.
-
Mull-Tokens: Modality-Agnostic Latent Thinking
Mull-Tokens are modality-agnostic latent tokens that enable free-form multimodal thinking and deliver up to 16% gains on spatial reasoning benchmarks.
-
OmniView-Space: Reinforcing Spatial Reasoning via Multi-Perspective Spatial Mapping
OmniView-Space framework with MPSM, tool-guided reasoning, and distillation achieves SOTA on spatial reasoning benchmarks for MLLMs while reducing external geometry dependencies.
-
SpatialSV: Internalizing Interpretable 3D Spatial Awareness in MLLMs via Task-Oriented Visual Supervision
SpatialSV trains MLLMs to lift 2D visual features into explicit 3D representations via task-oriented supervision for interpretable spatial awareness.
-
Thinking with Novel Views: A Systematic Analysis of Generative-Augmented Spatial Intelligence
Integrating generative novel-view synthesis into LMM reasoning loops improves accuracy on spatial subtasks by 1.3 to 3.9 percentage points across multiple models and tasks.
-
MAG-3D: Multi-Agent Grounded Reasoning for 3D Understanding
MAG-3D is a training-free multi-agent framework that coordinates planning, grounding, and coding agents with off-the-shelf VLMs to achieve grounded 3D reasoning and state-of-the-art benchmark results.
-
OpenSpatial: A Principled Data Engine for Empowering Spatial Intelligence
OpenSpatial supplies a principled open-source data engine and 3-million-sample dataset that raises spatial-reasoning model performance by an average of 19 percent on benchmarks.
-
EgoMind: Activating Spatial Cognition through Linguistic Reasoning in MLLMs
EgoMind uses Role-Play Caption and Progressive Spatial Analysis to give MLLMs competitive multi-frame spatial reasoning without 3D priors, using only 5K SFT and 20K RL samples.
-
OpenWorldLib: A Unified Codebase and Definition of Advanced World Models
OpenWorldLib defines world models as perception-centered systems with interaction and long-term memory, and provides a modular inference codebase unifying interactive video, 3D, reasoning, and VLA tasks.
-
Reinforcement Learning for Scalable and Trustworthy Intelligent Systems
Reinforcement learning is advanced for communication-efficient federated optimization and for preference-aligned, contextually safe policies in large language models.