# Deep Attention-guided Adaptive Subsampling

arXiv: https://arxiv.org/abs/2510.12376 · HTML: https://mlsanity.com/paper/2510.12376/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 10 references that are in the map.
- First posted on arXiv 14 Oct 2025.

## Builds on (most cited first, 7 of 7)

- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables](https://mlsanity.com/paper/1611.00712/) (2016)
- [MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification](https://mlsanity.com/paper/2110.14795/) (2021)
- [SCSampler: Sampling Salient Clips from Video for Efficient Action Recognition](https://mlsanity.com/paper/1904.04289/) (2019)
- [End-to-end Learning of Action Detection from Frame Glimpses in Videos](https://mlsanity.com/paper/1511.06984/) (2015)
- [A New Dataset and A Baseline Model for Breast Lesion Detection in Ultrasound Videos](https://mlsanity.com/paper/2207.00141/) (2022)

## Abstract (arXiv)

Although deep neural networks have provided impressive gains in performance, these improvements often come at the cost of increased computational complexity and expense. In many cases, such as 3D volume or video classification tasks, not all slices or frames are necessary due to inherent redundancies. To address this issue, we propose a novel learnable subsampling framework that can be integrated into any neural network architecture. Subsampling, being a nondifferentiable operation, poses significant challenges for direct adaptation into deep learning models. While some works, have proposed solutions using the Gumbel-max trick to overcome the problem of non-differentiability, they fall short in a crucial aspect: they are only task-adaptive and not inputadaptive. Once the sampling mechanism is learned, it remains static and does not adjust to different inputs, making it unsuitable for real-world applications. To this end, we propose an attention-guided sampling module that adapts to inputs even during inference. This dynamic adaptation results in performance gains and reduces complexity in deep neural network models. We demonstrate the effectiveness of our method on 3D medical imaging datasets from MedMNIST3D as well as two ultrasound video datasets for classification tasks, one of them being a challenging in-house dataset collected under real-world clinical conditions.