Towards the Connection between Activation Sparsity and Flat Minima

• URL: http://arxiv.org/abs/2605.25612v1
• Date: Mon, 25 May 2026 09:10:40 GMT
• Title: Towards the Connection between Activation Sparsity and Flat Minima
• Abstract summary: Activation sparsity emerges in blocks of standardly trained Transformers.<n>We find that activation sparsity equals a ratio between "augmented flatness" and the product of the input and activation norm.<n>We propose derivative sparsity, which reduces to activation sparsity under ReLU.
• Score: 46.83458699343789
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The observation that activation sparsity emerges in MLP blocks of standardly trained Transformers offers an opportunity to drastically reduce computation costs without sacrificing performance. To theoretically explain this phenomenon, existing works have shown that activation sparsity does not result from the data properties or data fitting but from the implicit bias of the training process. However, these connections are obtained with strong assumptions, which cannot be applied to deep models standardly trained with a large number of steps. Different from these works, we find that the flatness of loss landscapes is also closely related to the MLP activation sparsity and can serve as a weaker and naturally emerging assumption standard deep networks. Specifically, we find that 1) the MLP activation sparsity equals a ratio between "augmented flatness" (a weighted sum of flatness measures) and the product of the input norm and activation gradient of the MLP. We empirically find that this ratio decreases during training, leading to sparse activations. 2) We also propose the notion of derivative sparsity, which reduces to activation sparsity under ReLU, but further enables pruning in the backward propagation and is more stable than activation sparsity. With the theoretical findings, we can further encourage activation sparsity by decreasing the numerator and increasing the denominator of the ratio using three methods. These plug-and-play modifications can effectively reduce the ratio and produce sparser activations. Experiments on ImageNet-1K and C4 demonstrate relative improvements of at least 36% on inference sparsity and at least 50% on training sparsity over vanilla Transformers, indicating further potential cost reduction in both inference and training

Related papers

• Bug or Feature$^2$: Weight Drift, Activation Sparsity and Spikes [53.726365933748134]
  We analyze a negative weight drift induced by the interaction between standard losses and positively biased activation functions.<n>We characterize the sparsity-accuracy tradeoff across 79 configurations and identify a sharp accuracy cliff above $sim$70% activation sparsity.
  arXiv  Detail & Related papers  (2026-05-17T21:29:20Z)
• When Does Sparsity Mitigate the Curse of Depth in LLMs [53.137717161619484]
  We show that sparsity acts as a regulator of variance propagation and thereby improves depth utilization.<n>Our results reveal sparsity as a key yet previously overlooked mechanism for effective depth scaling in large language models.
  arXiv  Detail & Related papers  (2026-03-16T15:04:16Z)
• ActTail: Global Activation Sparsity in Large Language Models [14.120733678728191]
  We propose ActTail, a TopK magnitude-based activation sparsity method with global activation sparsity allocation grounded in Heavy-Tailed Self-Regularization (HT-SR) theory.<n>We show that our method improves both perplexity and downstream task performance at high sparsity compared to uniform allocation.
  arXiv  Detail & Related papers  (2026-02-18T14:46:03Z)
• Sparsing Law: Towards Large Language Models with Greater Activation Sparsity [64.15238674475619]
  Activation sparsity denotes the existence of substantial weakly-contributed elements within activation outputs that can be eliminated.<n>We propose PPL-$p%$ sparsity, a precise and performance-aware activation sparsity metric.<n>We show that ReLU is more efficient as the activation function than SiLU and can leverage more training data to improve activation sparsity.
  arXiv  Detail & Related papers  (2024-11-04T17:59:04Z)
• Learning Neural Networks with Sparse Activations [42.88109060676769]
  In transformer networks, the activations in the hidden layer of this block tend to be extremely sparse on any given input. Unlike traditional forms of sparsity, where there are neurons/weights which can be deleted from the network, this form of em activation sparsity appears to be harder to exploit. We present a variety of results showing that classes of functions do lead to provable computational and statistical advantages over their non-sparse counterparts.
  arXiv  Detail & Related papers  (2024-06-26T00:11:13Z)
• ProSparse: Introducing and Enhancing Intrinsic Activation Sparsity within Large Language Models [74.59731375779934]
  Activation sparsity refers to the existence of weakly-contributed elements among activation outputs.<n>This paper introduces a simple and effective sparsification method named "ProSparse" to push LLMs for higher activation sparsity.
  arXiv  Detail & Related papers  (2024-02-21T03:58:49Z)
• A Theoretical Explanation of Activation Sparsity through Flat Minima and Adversarial Robustness [29.87592869483743]
  A recent empirical observation of activation sparsity in blocks offers an opportunity to drastically reduce computation costs for free. We propose the notion of sparsity as one source of activation sparsity and a theoretical explanation based on it.
  arXiv  Detail & Related papers  (2023-09-06T13:48:40Z)
• The Lazy Neuron Phenomenon: On Emergence of Activation Sparsity in Transformers [59.87030906486969]
  This paper studies the curious phenomenon for machine learning models with Transformer architectures that their activation maps are sparse. We show that sparsity is a prevalent phenomenon that occurs for both natural language processing and vision tasks. We discuss how sparsity immediately implies a way to significantly reduce the FLOP count and improve efficiency for Transformers.
  arXiv  Detail & Related papers  (2022-10-12T15:25:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.