# Navigating Extremes: Dynamic Sparsity in Large Output Spaces

arXiv: https://arxiv.org/abs/2411.03171 · HTML: https://mlsanity.com/paper/2411.03171/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 51 of its 73 references that are in the map.
- First posted on arXiv 5 Nov 2024.

## Builds on (most cited first, 30 of 51)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://mlsanity.com/paper/1803.03635/) (2018)
- [Mixed Precision Training](https://mlsanity.com/paper/1710.03740/) (2017)
- [Training Deep Nets with Sublinear Memory Cost](https://mlsanity.com/paper/1604.06174/) (2016)
- [Linear Mode Connectivity and the Lottery Ticket Hypothesis](https://mlsanity.com/paper/1912.05671/) (2019)
- [Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networks](https://mlsanity.com/paper/2102.00554/) (2021)
- [Rigging the Lottery: Making All Tickets Winners](https://mlsanity.com/paper/1911.11134/) (2019)
- [Scalable Training of Artificial Neural Networks with Adaptive Sparse Connectivity inspired by Network Science](https://mlsanity.com/paper/1707.04780/) (2017)
- [Accelerating Sparse Deep Neural Networks](https://mlsanity.com/paper/2104.08378/) (2021)
- [Sparse Networks from Scratch: Faster Training without Losing Performance](https://mlsanity.com/paper/1907.04840/) (2019)
- [Proving the Lottery Ticket Hypothesis: Pruning is All You Need](https://mlsanity.com/paper/2002.00585/) (2020)
- [Sparse GPU Kernels for Deep Learning](https://mlsanity.com/paper/2006.10901/) (2020)
- [Dynamic Model Pruning with Feedback](https://mlsanity.com/paper/2006.07253/) (2020)
- [Do We Actually Need Dense Over-Parameterization? In-Time Over-Parameterization in Sparse Training](https://mlsanity.com/paper/2102.02887/) (2021)
- [AttentionXML: Label Tree-based Attention-Aware Deep Model for High-Performance Extreme Multi-Label Text Classification](https://mlsanity.com/paper/1811.01727/) (2018)
- [Inferring Networks of Substitutable and Complementary Products](https://mlsanity.com/paper/1506.08839/) (2015)
- [The Lottery Tickets Hypothesis for Supervised and Self-supervised Pre-training in Computer Vision Models](https://mlsanity.com/paper/2012.06908/) (2020)
- [Gradient Flow in Sparse Neural Networks and How Lottery Tickets Win](https://mlsanity.com/paper/2010.03533/) (2020)
- [Sparse DNNs with Improved Adversarial Robustness](https://mlsanity.com/paper/1810.09619/) (2018)
- [DiSMEC - Distributed Sparse Machines for Extreme Multi-label Classification](https://mlsanity.com/paper/1609.02521/) (2016)
- [Fast Multi-Resolution Transformer Fine-tuning for Extreme Multi-label Text Classification](https://mlsanity.com/paper/2110.00685/) (2021)
- [A Winning Hand: Compressing Deep Networks Can Improve Out-Of-Distribution Robustness](https://mlsanity.com/paper/2106.09129/) (2021)
- [Compressing LLMs: The Truth is Rarely Pure and Never Simple](https://mlsanity.com/paper/2310.01382/) (2023)
- [LightXML: Transformer with Dynamic Negative Sampling for High-Performance Extreme Multi-label Text Classification](https://mlsanity.com/paper/2101.03305/) (2021)
- [The State of Sparse Training in Deep Reinforcement Learning](https://mlsanity.com/paper/2206.10369/) (2022)
- [NeST: A Neural Network Synthesis Tool Based on a Grow-and-Prune Paradigm](https://mlsanity.com/paper/1711.02017/) (2017)
- [Deep Ensembling with No Overhead for either Training or Testing: The All-Round Blessings of Dynamic Sparsity](https://mlsanity.com/paper/2106.14568/) (2021)

## Built on by (most cited first, 3 of 3)

- [Activation Approximations Can Incur Safety Vulnerabilities Even in Aligned LLMs: Comprehensive Analysis and Defense](https://mlsanity.com/paper/2502.00840/) (2025)
- [Sparse CLIP: Co-Optimizing Interpretability and Performance in Contrastive Learning](https://mlsanity.com/paper/2601.20075/) (2026)
- [HASTE: Hardware-Aware Dynamic Sparse Training for Large Output Spaces](https://mlsanity.com/paper/2606.01117/) (2026)

## Abstract (arXiv)

In recent years, Dynamic Sparse Training (DST) has emerged as an alternative to post-training pruning for generating efficient models. In principle, DST allows for a more memory efficient training process, as it maintains sparsity throughout the entire training run. However, current DST implementations fail to capitalize on this in practice. Because sparse matrix multiplication is much less efficient than dense matrix multiplication on GPUs, most implementations simulate sparsity by masking weights. In this paper, we leverage recent advances in semi-structured sparse training to apply DST in the domain of classification with large output spaces, where memory-efficiency is paramount. With a label space of possibly millions of candidates, the classification layer alone will consume several gigabytes of memory. Switching from a dense to a fixed fan-in sparse layer updated with sparse evolutionary training (SET); however, severely hampers training convergence, especially at the largest label spaces. We find that poor gradient flow from the sparse classifier to the dense text encoder make it difficult to learn good input representations. By employing an intermediate layer or adding an auxiliary training objective, we recover most of the generalisation performance of the dense model. Overall, we demonstrate the applicability and practical benefits of DST in a challenging domain -- characterized by a highly skewed label distribution that differs substantially from typical DST benchmark datasets -- which enables end-to-end training with millions of labels on commodity hardware.