# Do VLMs Need Vision Transformers? Evaluating State Space Models as Vision Encoders

arXiv: https://arxiv.org/abs/2603.19209 · HTML: https://mlsanity.com/paper/2603.19209/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 41 references that are in the map.
- First posted on arXiv 19 Mar 2026.

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

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [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)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Towards VQA Models That Can Read](https://mlsanity.com/paper/1904.08920/) (2019)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [VMamba: Visual State Space Model](https://mlsanity.com/paper/2401.10166/) (2024)
- [A Survey on Multimodal Large Language Models](https://mlsanity.com/paper/2306.13549/) (2023)
- [Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs](https://mlsanity.com/paper/2406.16860/) (2024)
- [Exploring Plain Vision Transformer Backbones for Object Detection](https://mlsanity.com/paper/2203.16527/) (2022)
- [Vision Transformer Adapter for Dense Predictions](https://mlsanity.com/paper/2205.08534/) (2022)
- [MaxViT: Multi-Axis Vision Transformer](https://mlsanity.com/paper/2204.01697/) (2022)
- [Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Models](https://mlsanity.com/paper/2402.07865/) (2024)
- [SpatialRGPT: Grounded Spatial Reasoning in Vision Language Models](https://mlsanity.com/paper/2406.01584/) (2024)
- [MambaVision: A Hybrid Mamba-Transformer Vision Backbone](https://mlsanity.com/paper/2407.08083/) (2024)
- [TallyQA: Answering Complex Counting Questions](https://mlsanity.com/paper/1810.12440/) (2018)
- [Cobra: Extending Mamba to Multi-Modal Large Language Model for Efficient Inference](https://mlsanity.com/paper/2403.14520/) (2024)
- [Vision Transformers provably learn spatial structure](https://mlsanity.com/paper/2210.09221/) (2022)
- [VL-Mamba: Exploring State Space Models for Multimodal Learning](https://mlsanity.com/paper/2403.13600/) (2024)
- [Learning to Localize Objects Improves Spatial Reasoning in Visual-LLMs](https://mlsanity.com/paper/2404.07449/) (2024)
- [Spatial-Mamba: Effective Visual State Space Models via Structure-aware State Fusion](https://mlsanity.com/paper/2410.15091/) (2024)
- [2DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classification](https://mlsanity.com/paper/2412.00678/) (2024)
- [LLaVA-MORE: A Comparative Study of LLMs and Visual Backbones for Enhanced Visual Instruction Tuning](https://mlsanity.com/paper/2503.15621/) (2025)
- [Inference Optimal VLMs Need Fewer Visual Tokens and More Parameters](https://mlsanity.com/paper/2411.03312/) (2024)
- [DropPos: Pre-Training Vision Transformers by Reconstructing Dropped Positions](https://mlsanity.com/paper/2309.03576/) (2023)
- [OCID-Ref: A 3D Robotic Dataset with Embodied Language for Clutter Scene Grounding](https://mlsanity.com/paper/2103.07679/) (2021)
- [N"uwa: Mending the Spatial Integrity Torn by VLM Token Pruning](https://mlsanity.com/paper/2602.02951/) (2026)

## Abstract (arXiv)

Large vision--language models (VLMs) often use a frozen vision backbone, whose image features are mapped into a large language model through a lightweight connector. While transformer-based encoders are the standard visual backbone, we ask whether state space model (SSM) vision backbones can be a strong alternative. We systematically evaluate SSM vision backbones for VLMs in a controlled setting. Under matched ImageNet-1K initialization, the SSM backbone achieves the strongest overall performance across both VQA and grounding/localization. We further adapt both SSM and ViT-family backbones with detection or segmentation training and find that dense-task tuning generally improves performance across families; after this adaptation, the SSM backbone remains competitive while operating at a substantially smaller model scale. We further observe that (i) higher ImageNet accuracy or larger backbones do not reliably translate into better VLM performance, and (ii) some visual backbones are unstable in localization. Based on these findings, we propose stabilization strategies that improve robustness for both backbone families and highlight SSM backbones as a strong alternative to transformer-based vision encoders in VLMs.