# STAA: Spatio-Temporal Attention Attribution for Real-Time Interpreting Transformer-based Video Models

arXiv: https://arxiv.org/abs/2411.00630 · HTML: https://mlsanity.com/paper/2411.00630/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 37 references that are in the map.
- First posted on arXiv 1 Nov 2024.

## Builds on (most cited first, 22 of 22)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [Is Space-Time Attention All You Need for Video Understanding?](https://mlsanity.com/paper/2102.05095/) (2021)
- [ViViT: A Video Vision Transformer](https://mlsanity.com/paper/2103.15691/) (2021)
- [Video Swin Transformer](https://mlsanity.com/paper/2106.13230/) (2021)
- [Temporal Segment Networks: Towards Good Practices for Deep Action Recognition](https://mlsanity.com/paper/1608.00859/) (2016)
- [Transformer Interpretability Beyond Attention Visualization](https://mlsanity.com/paper/2012.09838/) (2020)
- [A Comprehensive Study of Deep Video Action Recognition](https://mlsanity.com/paper/2012.06567/) (2020)
- [Video Transformers: A Survey](https://mlsanity.com/paper/2201.05991/) (2022)
- [LoViT: Long Video Transformer for Surgical Phase Recognition](https://mlsanity.com/paper/2305.08989/) (2023)
- [Temporal-Channel Transformer for 3D Lidar-Based Video Object Detection in Autonomous Driving](https://mlsanity.com/paper/2011.13628/) (2020)
- [Don't Explain without Verifying Veracity: An Evaluation of Explainable AI with Video Activity Recognition](https://mlsanity.com/paper/2005.02335/) (2020)
- [Class Feature Pyramids for Video Explanation](https://mlsanity.com/paper/1909.08611/) (2019)
- [Discriminating Spatial and Temporal Relevance in Deep Taylor Decompositions for Explainable Activity Recognition](https://mlsanity.com/paper/1908.01536/) (2019)
- [XAIport: A Service Framework for the Early Adoption of XAI in AI Model Development](https://mlsanity.com/paper/2403.16858/) (2024)
- [Cloud-based XAI Services for Assessing Open Repository Models Under Adversarial Attacks](https://mlsanity.com/paper/2401.12261/) (2024)

## Built on by (most cited first, 2 of 2)

- [Keeping it Local, Tiny and Real: Automated Report Generation on Edge Computing Devices for Mechatronic-Based Cognitive Systems](https://mlsanity.com/paper/2511.02507/) (2025)
- [Interpreting Video Representations with Spatio-Temporal Sparse Autoencoders](https://mlsanity.com/paper/2604.03919/) (2026)

## Abstract (arXiv)

Transformer-based models have achieved state-of-the-art performance in various computer vision tasks, including image and video analysis. However, Transformer's complex architecture and black-box nature pose challenges for explainability, a crucial aspect for real-world applications and scientific inquiry. Current Explainable AI (XAI) methods can only provide one-dimensional feature importance, either spatial or temporal explanation, with significant computational complexity. This paper introduces STAA (Spatio-Temporal Attention Attribution), an XAI method for interpreting video Transformer models. Differ from traditional methods that separately apply image XAI techniques for spatial features or segment contribution analysis for temporal aspects, STAA offers both spatial and temporal information simultaneously from attention values in Transformers. The study utilizes the Kinetics-400 dataset, a benchmark collection of 400 human action classes used for action recognition research. We introduce metrics to quantify explanations. We also apply optimization to enhance STAA's raw output. By implementing dynamic thresholding and attention focusing mechanisms, we improve the signal-to-noise ratio in our explanations, resulting in more precise visualizations and better evaluation results. In terms of computational overhead, our method requires less than 3% of the computational resources of traditional XAI methods, making it suitable for real-time video XAI analysis applications. STAA contributes to the growing field of XAI by offering a method for researchers and practitioners to analyze Transformer models.