# Data augmentation techniques for the Video Question Answering task

arXiv: https://arxiv.org/abs/2008.09849 · HTML: https://mlsanity.com/paper/2008.09849/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Aug 2020.

## Built on by (most cited first, 5 of 5)

- [A Survey of Data Augmentation Approaches for NLP](https://mlsanity.com/paper/2105.03075/) (2021)
- [Video Question Answering: Datasets, Algorithms and Challenges](https://mlsanity.com/paper/2203.01225/) (2022)
- [Synthetic Human Action Video Data Generation with Pose Transfer](https://mlsanity.com/paper/2506.09411/) (2025)
- [Beyond Isolated Facts: Synthesizing Narrative and Grounded Supervision for VideoQA](https://mlsanity.com/paper/2509.24445/) (2025)
- [Evaluating Linguistic Capabilities of Multimodal LLMs in the Lens of Few-Shot Learning](https://mlsanity.com/paper/2407.12498/) (2024)

## Abstract (arXiv)

Video Question Answering (VideoQA) is a task that requires a model to analyze and understand both the visual content given by the input video and the textual part given by the question, and the interaction between them in order to produce a meaningful answer. In our work we focus on the Egocentric VideoQA task, which exploits first-person videos, because of the importance of such task which can have impact on many different fields, such as those pertaining the social assistance and the industrial training. Recently, an Egocentric VideoQA dataset, called EgoVQA, has been released. Given its small size, models tend to overfit quickly. To alleviate this problem, we propose several augmentation techniques which give us a +5.5% improvement on the final accuracy over the considered baseline.