# Behind the Scene: Revealing the Secrets of Pre-trained Vision-and-Language Models

arXiv: https://arxiv.org/abs/2005.07310 · HTML: https://mlsanity.com/paper/2005.07310/

## Key facts

- Cited by 90 papers in the map (data through 7 Oct 2026).
- First posted on arXiv 15 May 2020.

## Built on by (most cited first, 30 of 90)

- [Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts](https://mlsanity.com/paper/2102.08981/) (2021)
- [Winoground: Probing Vision and Language Models for Visio-Linguistic Compositionality](https://mlsanity.com/paper/2204.03162/) (2022)
- [Multimodal Learning with Transformers: A Survey](https://mlsanity.com/paper/2206.06488/) (2022)
- [History Aware Multimodal Transformer for Vision-and-Language Navigation](https://mlsanity.com/paper/2110.13309/) (2021)
- [Scaling Up Vision-Language Pre-training for Image Captioning](https://mlsanity.com/paper/2111.12233/) (2021)
- [Devils in Middle Layers of Large Vision-Language Models: Interpreting, Detecting and Mitigating Object Hallucinations via Attention Lens](https://mlsanity.com/paper/2411.16724/) (2024)
- [VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena](https://mlsanity.com/paper/2112.07566/) (2021)
- [Vision-Language Pre-training: Basics, Recent Advances, and Future Trends](https://mlsanity.com/paper/2210.09263/) (2022)
- [Foundations and Trends in Multimodal Machine Learning: Principles, Challenges, and Open Questions](https://mlsanity.com/paper/2209.03430/) (2022)
- [Probing Image-Language Transformers for Verb Understanding](https://mlsanity.com/paper/2106.09141/) (2021)
- [A Good Prompt Is Worth Millions of Parameters: Low-resource Prompt-based Learning for Vision-Language Models](https://mlsanity.com/paper/2110.08484/) (2021)
- [TerraMind: Large-Scale Generative Multimodality for Earth Observation](https://mlsanity.com/paper/2504.11171/) (2025)
- [Characterizing and overcoming the greedy nature of learning in multi-modal deep neural networks](https://mlsanity.com/paper/2202.05306/) (2022)
- [Reliable Visual Question Answering: Abstain Rather Than Answer Incorrectly](https://mlsanity.com/paper/2204.13631/) (2022)
- [Compressing Visual-linguistic Model via Knowledge Distillation](https://mlsanity.com/paper/2104.02096/) (2021)
- [Adversarial VQA: A New Benchmark for Evaluating the Robustness of VQA Models](https://mlsanity.com/paper/2106.00245/) (2021)
- [Vision-Language-Action (VLA) Models: Concepts, Progress, Applications and Challenges](https://mlsanity.com/paper/2505.04769/) (2025)
- [Vision-and-Language or Vision-for-Language? On Cross-Modal Influence in Multimodal Transformers](https://mlsanity.com/paper/2109.04448/) (2021)
- [Probing Inter-modality: Visual Parsing with Self-Attention for Vision-Language Pre-training](https://mlsanity.com/paper/2106.13488/) (2021)
- [Cross-modal Information Flow in Multimodal Large Language Models](https://mlsanity.com/paper/2411.18620/) (2024)
- [Decoupling the Role of Data, Attention, and Losses in Multimodal Transformers](https://mlsanity.com/paper/2102.00529/) (2021)
- [Towards Vision-Language Mechanistic Interpretability: A Causal Tracing Tool for BLIP](https://mlsanity.com/paper/2308.14179/) (2023)
- [The Road to Know-Where: An Object-and-Room Informed Sequential BERT for Indoor Vision-Language Navigation](https://mlsanity.com/paper/2104.04167/) (2021)
- [Multi-Object Hallucination in Vision-Language Models](https://mlsanity.com/paper/2407.06192/) (2024)
- [EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive Pruning](https://mlsanity.com/paper/2210.07795/) (2022)
- [VL-InterpreT: An Interactive Visualization Tool for Interpreting Vision-Language Transformers](https://mlsanity.com/paper/2203.17247/) (2022)
- [Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2412.02104/) (2024)
- [Playing Lottery Tickets with Vision and Language](https://mlsanity.com/paper/2104.11832/) (2021)
- [VLMixer: Unpaired Vision-Language Pre-training via Cross-Modal CutMix](https://mlsanity.com/paper/2206.08919/) (2022)
- [MM-SHAP: A Performance-agnostic Metric for Measuring Multimodal Contributions in Vision and Language Models & Tasks](https://mlsanity.com/paper/2212.08158/) (2022)

## Abstract (arXiv)

Recent Transformer-based large-scale pre-trained models have revolutionized vision-and-language (V+L) research. Models such as ViLBERT, LXMERT and UNITER have significantly lifted state of the art across a wide range of V+L benchmarks with joint image-text pre-training. However, little is known about the inner mechanisms that destine their impressive success. To reveal the secrets behind the scene of these powerful models, we present VALUE (Vision-And-Language Understanding Evaluation), a set of meticulously designed probing tasks (e.g., Visual Coreference Resolution, Visual Relation Detection, Linguistic Probing Tasks) generalizable to standard pre-trained V+L models, aiming to decipher the inner workings of multimodal pre-training (e.g., the implicit knowledge garnered in individual attention heads, the inherent cross-modal alignment learned through contextualized multimodal embeddings). Through extensive analysis of each archetypal model architecture via these probing tasks, our key observations are: (i) Pre-trained models exhibit a propensity for attending over text rather than images during inference. (ii) There exists a subset of attention heads that are tailored for capturing cross-modal interactions. (iii) Learned attention matrix in pre-trained models demonstrates patterns coherent with the latent alignment between image regions and textual words. (iv) Plotted attention patterns reveal visually-interpretable relations among image regions. (v) Pure linguistic knowledge is also effectively encoded in the attention heads. These are valuable insights serving to guide future work towards designing better model architecture and objectives for multimodal pre-training.