# MLP Architectures for Vision-and-Language Modeling: An Empirical Study

arXiv: https://arxiv.org/abs/2112.04453 · HTML: https://mlsanity.com/paper/2112.04453/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 61 of its 70 references that are in the map.
- First posted on arXiv 8 Dec 2021.

## Builds on (most cited first, 30 of 61)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [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)
- [ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks](https://mlsanity.com/paper/1908.02265/) (2019)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [LXMERT: Learning Cross-Modality Encoder Representations from Transformers](https://mlsanity.com/paper/1908.07490/) (2019)
- [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021)
- [Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models](https://mlsanity.com/paper/1505.04870/) (2015)
- [Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering](https://mlsanity.com/paper/1707.07998/) (2017)
- [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://mlsanity.com/paper/1908.03557/) (2019)
- [ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision](https://mlsanity.com/paper/2102.03334/) (2021)
- [Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks](https://mlsanity.com/paper/2004.06165/) (2020)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [Deep Visual-Semantic Alignments for Generating Image Descriptions](https://mlsanity.com/paper/1412.2306/) (2014)
- [VL-BERT: Pre-training of Generic Visual-Linguistic Representations](https://mlsanity.com/paper/1908.08530/) (2019)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [Unified Vision-Language Pre-Training for Image Captioning and VQA](https://mlsanity.com/paper/1909.11059/) (2019)
- [Less is More: ClipBERT for Video-and-Language Learning via Sparse Sampling](https://mlsanity.com/paper/2102.06183/) (2021)
- [A Corpus for Reasoning About Natural Language Grounded in Photographs](https://mlsanity.com/paper/1811.00491/) (2018)
- [Dynabench: Rethinking Benchmarking in NLP](https://mlsanity.com/paper/2104.14337/) (2021)

## Built on by (most cited first, 6 of 6)

- [Vision-Language Pre-training: Basics, Recent Advances, and Future Trends](https://mlsanity.com/paper/2210.09263/) (2022)
- [Vision-Language-Action (VLA) Models: Concepts, Progress, Applications and Challenges](https://mlsanity.com/paper/2505.04769/) (2025)
- [TVLT: Textless Vision-Language Transformer](https://mlsanity.com/paper/2209.14156/) (2022)
- [AiM: Taking Answers in Mind to Correct Chinese Cloze Tests in Educational Applications](https://mlsanity.com/paper/2208.12505/) (2022)
- [Systematic Characterization of Minimal Deep Learning Architectures: A Unified Analysis of Convergence, Pruning, and Quantization](https://mlsanity.com/paper/2601.17987/) (2026)
- [CoMLP: Cooperatively-Gated MLPs for Fine-Grained Cross-Modal Information Fusion in Medical Image Segmentation](https://mlsanity.com/paper/2609.04781/) (2026)

## Abstract (arXiv)

We initiate the first empirical study on the use of MLP architectures for vision-and-language (VL) fusion. Through extensive experiments on 5 VL tasks and 5 robust VQA benchmarks, we find that: (i) Without pre-training, using MLPs for multimodal fusion has a noticeable performance gap compared to transformers; (ii) However, VL pre-training can help close the performance gap; (iii) Instead of heavy multi-head attention, adding tiny one-head attention to MLPs is sufficient to achieve comparable performance to transformers. Moreover, we also find that the performance gap between MLPs and transformers is not widened when being evaluated on the harder robust VQA benchmarks, suggesting using MLPs for VL fusion can generalize roughly to a similar degree as using transformers. These results hint that MLPs can effectively learn to align vision and text features extracted from lower-level encoders without heavy reliance on self-attention. Based on this, we ask an even bolder question: can we have an all-MLP architecture for VL modeling, where both VL fusion and the vision encoder are replaced with MLPs? Our result shows that an all-MLP VL model is sub-optimal compared to state-of-the-art full-featured VL models when both of them get pre-trained. However, pre-training an all-MLP can surprisingly achieve a better average score than full-featured transformer models without pre-training. This indicates the potential of large-scale pre-training of MLP-like architectures for VL modeling and inspires the future research direction on simplifying well-established VL modeling with less inductive design bias. Our code is publicly available at: https://github.com/easonnie/mlp-vil