# Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models

arXiv: https://arxiv.org/abs/2503.17349 · HTML: https://mlsanity.com/paper/2503.17349/

## Key facts

- Cited by 37 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 35 of its 43 references that are in the map.
- First posted on arXiv 21 Mar 2025.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)

## Builds on (most cited first, 30 of 35)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models](https://mlsanity.com/paper/2304.10592/) (2023)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [VizWiz Grand Challenge: Answering Visual Questions from Blind People](https://mlsanity.com/paper/1802.08218/) (2018)
- [Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs](https://mlsanity.com/paper/2406.16860/) (2024)
- [SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities](https://mlsanity.com/paper/2401.12168/) (2024)
- [An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models](https://mlsanity.com/paper/2403.06764/) (2024)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [SpatialRGPT: Grounded Spatial Reasoning in Vision Language Models](https://mlsanity.com/paper/2406.01584/) (2024)
- [What's "up" with vision-language models? Investigating their struggle with spatial reasoning](https://mlsanity.com/paper/2310.19785/) (2023)
- [Scaling Up Vision-Language Pre-training for Image Captioning](https://mlsanity.com/paper/2111.12233/) (2021)
- [Linear Representations of Sentiment in Large Language Models](https://mlsanity.com/paper/2310.15154/) (2023)
- [Birth of a Transformer: A Memory Viewpoint](https://mlsanity.com/paper/2306.00802/) (2023)
- [AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmark](https://mlsanity.com/paper/2410.03051/) (2024)
- [Towards Interpreting Visual Information Processing in Vision-Language Models](https://mlsanity.com/paper/2410.07149/) (2024)
- [From CLIP to DINO: Visual Encoders Shout in Multi-modal Large Language Models](https://mlsanity.com/paper/2310.08825/) (2023)

## Built on by (most cited first, 30 of 37)

- [Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence](https://mlsanity.com/paper/2505.23747/) (2025)
- [Visual Representation Alignment for Multimodal Large Language Models](https://mlsanity.com/paper/2509.07979/) (2025)
- [Video-MMLU: A Massive Multi-Discipline Lecture Understanding Benchmark](https://mlsanity.com/paper/2504.14693/) (2025)
- [Multimodal Spatial Reasoning in the Large Model Era: A Survey and Benchmarks](https://mlsanity.com/paper/2510.25760/) (2025)
- [Thinking with Blueprints: Assisting Vision-Language Models in Spatial Reasoning via Structured Object Representation](https://mlsanity.com/paper/2601.01984/) (2026)
- [Large Language Model Reasoning Failures](https://mlsanity.com/paper/2602.06176/) (2026)
- [SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning](https://mlsanity.com/paper/2603.27437/) (2026)
- [CompoDistill: Attention Distillation for Compositional Reasoning in Multimodal LLMs](https://mlsanity.com/paper/2510.12184/) (2025)
- [Rethinking Causal Mask Attention for Vision-Language Inference](https://mlsanity.com/paper/2505.18605/) (2025)
- [The Unseen Bias: How Norm Discrepancy in Pre-Norm MLLMs Leads to Visual Information Loss](https://mlsanity.com/paper/2512.08374/) (2025)
- [Attention in Space: Functional Roles of VLM Heads for Spatial Reasoning](https://mlsanity.com/paper/2603.20662/) (2026)
- [World2VLM: Distilling World Model Imagination into VLMs for Dynamic Spatial Reasoning](https://mlsanity.com/paper/2604.26934/) (2026)
- [CubeBench: Diagnosing Interactive, Long-Horizon Spatial Reasoning Under Partial Observations](https://mlsanity.com/paper/2512.23328/) (2025)
- [VIEW2SPACE: Studying Multi-View Visual Reasoning from Sparse Observations](https://mlsanity.com/paper/2603.16506/) (2026)
- [Beyond the Vision Encoder: Identifying and Mitigating Spatial Bias in Large Vision-Language Models](https://mlsanity.com/paper/2509.21984/) (2025)
- [Mitigating Coordinate Prediction Bias from Positional Encoding Failures](https://mlsanity.com/paper/2510.22102/) (2025)
- [Unleashing the Intrinsic Visual Representation Capability of Multimodal Large Language Models](https://mlsanity.com/paper/2512.06281/) (2025)
- [3D Primitives are a Spatial Language for VLMs](https://mlsanity.com/paper/2605.12586/) (2026)
- [See What I Mean: Aligning Vision and Language Representations for Video Fine-grained Object Understanding](https://mlsanity.com/paper/2605.18018/) (2026)
- [AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Models](https://mlsanity.com/paper/2606.08952/) (2026)
- [Decodable Is Not Grounded: A Vision-Ablation Arbiter for VLM Spatial Reasoning](https://mlsanity.com/paper/2606.31257/) (2026)
- [Seeing Clearly, Reasoning Confidently: Plug-and-Play Remedies for Vision Language Model Blindness](https://mlsanity.com/paper/2602.19615/) (2026)
- [World2Mind: Cognition Toolkit for Allocentric Spatial Reasoning in Foundation Models](https://mlsanity.com/paper/2603.09774/) (2026)
- [Why Do Vision Language Models Struggle To Recognize Human Emotions?](https://mlsanity.com/paper/2604.15280/) (2026)
- [HyCal: A Training-Free Prototype Calibration Method for Cross-Discipline Few-Shot Class-Incremental Learning](https://mlsanity.com/paper/2604.15678/) (2026)
- [Uncovering and Shaping the Latent Representation of 3D Scene Topology in Vision-Language Models](https://mlsanity.com/paper/2605.07148/) (2026)
- [Where, Not What: Compelling Video LLMs to Learn Geometric Causality for 3D-Grounding](https://mlsanity.com/paper/2510.17034/) (2025)
- [A Multi-Modal Neuro-Symbolic Approach for Spatial Reasoning-Based Visual Grounding in Robotics](https://mlsanity.com/paper/2510.27033/) (2025)
- [An Evaluation of Interleaved Instruction Tuning on Semantic Reasoning Performance in an Audio MLLM](https://mlsanity.com/paper/2511.02234/) (2025)
- [The Cost of Language: Centroid Erasure Exposes and Exploits Modal Competition in Multimodal Language Models](https://mlsanity.com/paper/2604.14363/) (2026)

## Abstract (arXiv)

Vision Language Models (VLMs) excel at identifying and describing objects but often fail at spatial reasoning. We study why VLMs, such as LLaVA, underutilize spatial cues despite having positional encodings and spatially rich vision encoder features. Our analysis reveals a key imbalance: vision token embeddings have much larger norms than text tokens, suppressing LLM's position embedding. To expose this mechanism, we developed three interpretability tools: (1) the Position Sensitivity Index, which quantifies reliance on token order, (2) the Cross Modality Balance, which reveals attention head allocation patterns, and (3) a RoPE Sensitivity probe, which measures dependence on rotary positional embeddings. These tools uncover that vision tokens and system prompts dominate attention. We validated our mechanistic understanding through targeted interventions that predictably restore positional sensitivity. These findings reveal previously unknown failure modes in multimodal attention and demonstrate how interpretability analysis can guide principled improvements.