# Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding

arXiv: https://arxiv.org/abs/2503.06287 · HTML: https://mlsanity.com/paper/2503.06287/

## Key facts

- Cited by 83 papers in the map (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 70 of its 82 references that are in the map.
- First posted on arXiv 8 Mar 2025.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 30 of 70)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [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)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond](https://mlsanity.com/paper/2308.12966/) (2023)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Modeling Context in Referring Expressions](https://mlsanity.com/paper/1608.00272/) (2016)
- [ShareGPT4V: Improving Large Multi-Modal Models with Better Captions](https://mlsanity.com/paper/2311.12793/) (2023)
- [What Does BERT Look At? An Analysis of BERT's Attention](https://mlsanity.com/paper/1906.04341/) (2019)
- [Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic](https://mlsanity.com/paper/2306.15195/) (2023)
- [Generation and Comprehension of Unambiguous Object Descriptions](https://mlsanity.com/paper/1511.02283/) (2015)
- [Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned](https://mlsanity.com/paper/1905.09418/) (2019)
- [Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs](https://mlsanity.com/paper/2406.16860/) (2024)
- [MDETR -- Modulated Detection for End-to-End Multi-Modal Understanding](https://mlsanity.com/paper/2104.12763/) (2021)
- [LISA: Reasoning Segmentation via Large Language Model](https://mlsanity.com/paper/2308.00692/) (2023)
- [DeepSeek-VL: Towards Real-World Vision-Language Understanding](https://mlsanity.com/paper/2403.05525/) (2024)
- [MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing](https://mlsanity.com/paper/2304.08465/) (2023)
- [Yi: Open Foundation Models by 01.AI](https://mlsanity.com/paper/2403.04652/) (2024)
- [CogVLM: Visual Expert for Pretrained Language Models](https://mlsanity.com/paper/2311.03079/) (2023)
- [Ferret: Refer and Ground Anything Anywhere at Any Granularity](https://mlsanity.com/paper/2310.07704/) (2023)
- [GLaMM: Pixel Grounding Large Multimodal Model](https://mlsanity.com/paper/2311.03356/) (2023)
- [Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models](https://mlsanity.com/paper/2403.18814/) (2024)

## Built on by (most cited first, 30 of 83)

- [Image Generators are Generalist Vision Learners](https://mlsanity.com/paper/2604.20329/) (2026)
- [Visual Representation Alignment for Multimodal Large Language Models](https://mlsanity.com/paper/2509.07979/) (2025)
- [Towards Cross-View Point Correspondence in Vision-Language Models](https://mlsanity.com/paper/2512.04686/) (2025)
- [Constructive Distortion: Improving MLLMs with Attention-Guided Image Warping](https://mlsanity.com/paper/2510.09741/) (2025)
- [LatBot: Distilling Universal Latent Actions for Vision-Language-Action Models](https://mlsanity.com/paper/2511.23034/) (2025)
- [How Do Large Vision-Language Models See Text in Image? Unveiling the Distinctive Role of OCR Heads](https://mlsanity.com/paper/2505.15865/) (2025)
- [Unleashing the Potential of Multimodal LLMs for Zero-Shot Spatio-Temporal Video Grounding](https://mlsanity.com/paper/2509.15178/) (2025)
- [Catching the Details: Self-Distilled RoI Predictors for Fine-Grained MLLM Perception](https://mlsanity.com/paper/2509.16944/) (2025)
- [UniVLR: Unifying Text and Vision in Visual Latent Reasoning for Multimodal LLMs](https://mlsanity.com/paper/2605.11856/) (2026)
- [Direct Visual Grounding by Directing Attention of Visual Tokens](https://mlsanity.com/paper/2511.12738/) (2025)
- [Q-Zoom: Query-Aware Adaptive Perception for Efficient Multimodal Large Language Models](https://mlsanity.com/paper/2604.06912/) (2026)
- [ReFineVLA: Reasoning-Aware Teacher-Guided Transfer Fine-Tuning](https://mlsanity.com/paper/2505.19080/) (2025)
- [Seeing Beyond Words: Self-Supervised Visual Learning for Multimodal Large Language Models](https://mlsanity.com/paper/2512.15885/) (2025)
- [Your Vision-Language-Action Model Already Has Attention Heads For Path Deviation Detection](https://mlsanity.com/paper/2603.13782/) (2026)
- [Gaze Heads: How VLMs Look at What They Describe](https://mlsanity.com/paper/2606.14703/) (2026)
- [Tracing and Mitigating Hallucinations in Multimodal LLMs via Dynamic Attention Localization](https://mlsanity.com/paper/2509.07864/) (2025)
- [Reallocating Attention Across Layers to Reduce Multimodal Hallucination](https://mlsanity.com/paper/2510.10285/) (2025)
- [Attention Misses Visual Risk: Risk-Adaptive Steering for Multimodal Safety Alignment](https://mlsanity.com/paper/2510.13698/) (2025)
- [Taming SAM3 in the Wild: A Concept Bank for Open-Vocabulary Segmentation](https://mlsanity.com/paper/2602.06333/) (2026)
- [Seeing Clearly without Training: Mitigating Hallucinations in Multimodal LLMs for Remote Sensing](https://mlsanity.com/paper/2603.02754/) (2026)
- [Mechanistic Insights into Functional Sparsity in Multimodal LLMs via CoRe Heads](https://mlsanity.com/paper/2606.05843/) (2026)
- [Your Model Already Knows: Attention-Guided Safety Filter for Vision-Language-Action Models](https://mlsanity.com/paper/2606.09749/) (2026)
- [RxnCaption: Reformulating Reaction Diagram Parsing as Visual Prompt Guided Captioning](https://mlsanity.com/paper/2511.02384/) (2025)
- [When to Think and When to Look: Uncertainty-Guided Lookback](https://mlsanity.com/paper/2511.15613/) (2025)
- [RGBT-GroundBench: Visual Grounding Beyond RGB in Complex Real-World Scenarios](https://mlsanity.com/paper/2512.24561/) (2025)
- [PEARL: Geometry Aligns Semantics for Training-Free Open-Vocabulary Semantic Segmentation](https://mlsanity.com/paper/2603.21528/) (2026)
- [Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations](https://mlsanity.com/paper/2604.04863/) (2026)
- [Preference Redirection via Attention Concentration: An Attack on Computer Use Agents](https://mlsanity.com/paper/2604.08005/) (2026)
- [Entropy-Gradient Grounding: Training-Free Evidence Retrieval in Vision-Language Models](https://mlsanity.com/paper/2604.08456/) (2026)
- [Aligning What Vision-Language Models See and Perceive with Adaptive Information Flow](https://mlsanity.com/paper/2604.15809/) (2026)

## Abstract (arXiv)

Visual grounding seeks to localize the image region corresponding to a free-form text description. Recently, the strong multimodal capabilities of Large Vision-Language Models (LVLMs) have driven substantial improvements in visual grounding, though they inevitably require fine-tuning and additional model components to explicitly generate bounding boxes or segmentation masks. However, we discover that a few attention heads in frozen LVLMs demonstrate strong visual grounding capabilities. We refer to these heads, which consistently capture object locations related to text semantics, as localization heads. Using localization heads, we introduce a straightforward and effective training-free visual grounding framework that utilizes text-to-image attention maps from localization heads to identify the target objects. Surprisingly, only three out of thousands of attention heads are sufficient to achieve competitive localization performance compared to existing LVLM-based visual grounding methods that require fine-tuning. Our findings suggest that LVLMs can innately ground objects based on a deep comprehension of the text-image relationship, as they implicitly focus on relevant image regions to generate informative text outputs. All the source codes will be made available to the public.