# ASSS: A Differentiable Adversarial Framework for Task-Aware Data Reduction

arXiv: https://arxiv.org/abs/2601.02081 · HTML: https://mlsanity.com/paper/2601.02081/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 25 references that are in the map.
- First posted on arXiv 5 Jan 2026.

## Builds on (most cited first, 15 of 15)

- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [A Survey of Deep Active Learning](https://mlsanity.com/paper/2009.00236/) (2020)
- [GRAD-MATCH: Gradient Matching based Data Subset Selection for Efficient Deep Model Training](https://mlsanity.com/paper/2103.00123/) (2021)
- [GLISTER: Generalization based Data Subset Selection for Efficient and Robust Learning](https://mlsanity.com/paper/2012.10630/) (2020)
- [Data Distillation: A Survey](https://mlsanity.com/paper/2301.04272/) (2023)
- [Practical Coreset Constructions for Machine Learning](https://mlsanity.com/paper/1703.06476/) (2017)
- [Data-Efficient Learning via Clustering-Based Sensitivity Sampling: Foundation Models and Beyond](https://mlsanity.com/paper/2402.17327/) (2024)
- [Gradient-based Sampling: An Adaptive Importance Sampling for Least-squares](https://mlsanity.com/paper/1803.00841/) (2018)
- [Advancing Deep Active Learning & Data Subset Selection: Unifying Principles with Information-Theory Intuitions](https://mlsanity.com/paper/2401.04305/) (2024)
- [Training Data Subset Search with Ensemble Active Learning](https://mlsanity.com/paper/1905.12737/) (2019)
- [Diversity Subsampling: Custom Subsamples from Large Data Sets](https://mlsanity.com/paper/2206.10812/) (2022)
- [Adapted variable density subsampling for compressed sensing](https://mlsanity.com/paper/2206.13796/) (2022)
- [What is the Right Notion of Distance between Predict-then-Optimize Tasks?](https://mlsanity.com/paper/2409.06997/) (2024)
- [Recurrent event analysis in the presence of real-time high frequency data via random subsampling](https://mlsanity.com/paper/2204.06632/) (2022)
- [Tackling the subsampling problem to infer collective properties from limited data](https://mlsanity.com/paper/2209.05548/) (2022)

## Abstract (arXiv)

Massive datasets often contain redundancy that inflates computational costs without improving generalization. Existing data reduction methods are typically task-agnostic, discarding informative boundary samples and yielding suboptimal performance. We propose Adversarial Soft-Selection Subsampling (ASSS), a differentiable framework that casts data reduction as a minimax game between a learnable selector and a task network. Using Gumbel-Softmax relaxation, ASSS enables end-to-end gradient flow and is theoretically grounded in the information bottleneck principle. Experiments on multiple benchmarks show that ASSS achieves a performance retention rate (PRR) of 98.9% while using only 30% of the data, significantly outperforming random sampling, K-means, and gradient-based methods. Visualizations confirm that ASSS preferentially retains samples near decision boundaries. The framework is scalable, fully differentiable, and easily integrated into existing training pipelines. This work introduces a new paradigm for task-aware data reduction that directly optimizes subset selection for the downstream objective, offering a principled and practical solution to the scalability challenges in modern deep learning.