OmniCoT is a new panoramic reasoning benchmark with 6.7K eval, 1K real, and 14.3K training examples plus a two-stage SFT+GRPO training method to enforce global 360-degree consistency.
SpatialBench: Benchmarking Multimodal Large Language Models for Spatial Cognition
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
abstract
Spatial cognition is fundamental to real-world multimodal intelligence, allowing models to effectively interact with the physical environment. While multimodal large language models (MLLMs) have made significant strides, existing benchmarks often oversimplify spatial cognition, reducing it to a single-dimensional metric, which fails to capture the hierarchical structure and interdependence of spatial abilities. To address this gap, we propose a hierarchical spatial cognition framework that decomposes spatial intelligence into five progressively complex levels from basic observation to high-level planning. Building upon this taxonomy, we construct SpatialBench, a large-scale, fine-grained benchmark covering 15 tasks aligned with these cognitive levels. To provide a unified evaluation across heterogeneous tasks, we further introduce a high-level capability-oriented metric that reliably assesses a model's overall spatial reasoning ability. Extensive experiments over massive MLLMs reveal distinct performance stratification across cognitive levels: models exhibit strong perceptual grounding yet remain limited in symbolic reasoning, causal inference, and planning. Additional human tests demonstrate that humans perform selective, goal-directed abstraction, while MLLMs tend to over-attend to surface details without coherent spatial intent. Our work establishes the first systematic framework for measuring hierarchical spatial cognition in MLLMs, laying the foundation for future spatially intelligent systems.
citation-role summary
background 2
dataset 1
citation-polarity summary
years
2026 11representative citing papers
A new diagnostic benchmark decomposes LLM spatial navigation into three cognitive scales and shows that cross-scale aggregation, not single-level deficits, causes failure beyond small mazes.
SpatialWorld is a new multi-simulator benchmark showing top multimodal agents achieve under 18% success on interactive spatial tasks requiring active exploration and long-horizon planning.
OVO-S-Bench provides 1680 human-annotated questions on 348 videos to measure streaming spatial intelligence in MLLMs across instantaneous perception, spatiotemporal tracking, spatial simulation, and allocentric mapping.
Frontier VLMs overconfidently answer spatial questions under occlusion (~30% accuracy) and perspective ambiguity (<10% accuracy) instead of abstaining, and often fail to select helpful additional views.
3D geometric primitives in executable code act as an effective intermediate spatial language that boosts VLMs on reconstruction and question-answering tasks.
ARGOS is the first benchmark reformulating multi-camera person search as an agentic interactive reasoning task grounded in a spatio-temporal topology graph, with 2691 tasks across three tracks where current LLMs achieve low accuracy.
TableVision benchmark shows explicit spatial grounding recovers MLLM reasoning on hierarchical tables, delivering 12.3% accuracy improvement through a decoupled perception-reasoning framework.
Distills 3D spatial reasoning from a 7B teacher VLM to a 2.29B student using VGGT encoder, multi-task losses, and Hidden CoT latent tokens, yielding 8.7x lower latency with 54-72% performance retention on ScanNet and 3D-FRONT.
VLMs possess a latent 3D scene topology subspace corresponding to Laplacian eigenmaps that can be causally shaped via Dirichlet energy regularization to improve spatial task performance by up to 12.1%.
SocialGrid benchmark shows even top LLMs achieve below 60% in embodied planning and task completion, with deception detection near random chance regardless of model scale.
citing papers explorer
-
OmniCoT: A Benchmark for Global and Multi-Step Panoramic Reasoning
OmniCoT is a new panoramic reasoning benchmark with 6.7K eval, 1K real, and 14.3K training examples plus a two-stage SFT+GRPO training method to enforce global 360-degree consistency.
-
Lost in Aggregation: A Multi-Scale Diagnostic Benchmark for LLM Spatial Navigation
A new diagnostic benchmark decomposes LLM spatial navigation into three cognitive scales and shows that cross-scale aggregation, not single-level deficits, causes failure beyond small mazes.
-
SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks
SpatialWorld is a new multi-simulator benchmark showing top multimodal agents achieve under 18% success on interactive spatial tasks requiring active exploration and long-horizon planning.
-
OVO-S-Bench: A Hierarchical Benchmark for Streaming Spatial Intelligence in Multimodal LLMs
OVO-S-Bench provides 1680 human-annotated questions on 348 videos to measure streaming spatial intelligence in MLLMs across instantaneous perception, spatiotemporal tracking, spatial simulation, and allocentric mapping.
-
Seeing Isn't Knowing: Do VLMs Know When Not to Answer Spatial Questions (and Why)?
Frontier VLMs overconfidently answer spatial questions under occlusion (~30% accuracy) and perspective ambiguity (<10% accuracy) instead of abstaining, and often fail to select helpful additional views.
-
3D Primitives are a Spatial Language for VLMs
3D geometric primitives in executable code act as an effective intermediate spatial language that boosts VLMs on reconstruction and question-answering tasks.
-
ARGOS: Who, Where, and When in Agentic Multi-Camera Person Search
ARGOS is the first benchmark reformulating multi-camera person search as an agentic interactive reasoning task grounded in a spatio-temporal topology graph, with 2691 tasks across three tracks where current LLMs achieve low accuracy.
-
TableVision: A Large-Scale Benchmark for Spatially Grounded Reasoning over Complex Hierarchical Tables
TableVision benchmark shows explicit spatial grounding recovers MLLM reasoning on hierarchical tables, delivering 12.3% accuracy improvement through a decoupled perception-reasoning framework.
-
Distilling 3D Spatial Reasoning into a Lightweight Vision-Language Model with CoT
Distills 3D spatial reasoning from a 7B teacher VLM to a 2.29B student using VGGT encoder, multi-task losses, and Hidden CoT latent tokens, yielding 8.7x lower latency with 54-72% performance retention on ScanNet and 3D-FRONT.
-
Uncovering and Shaping the Latent Representation of 3D Scene Topology in Vision-Language Models
VLMs possess a latent 3D scene topology subspace corresponding to Laplacian eigenmaps that can be causally shaped via Dirichlet energy regularization to improve spatial task performance by up to 12.1%.
-
SocialGrid: A Benchmark for Planning and Social Reasoning in Embodied Multi-Agent Systems
SocialGrid benchmark shows even top LLMs achieve below 60% in embodied planning and task completion, with deception detection near random chance regardless of model scale.