paper-with-me

Papers

LEAP: Learnable Pruning for Transformer-based Models

2021-05-30 · Zhewei Yao, Xiaoxia Wu, Linjian Ma, Sheng Shen, Kurt Keutzer, Michael W. Mahoney, Yuxiong He

Pruning is an effective method to reduce the memory footprint and computational cost associated with large natural language processing models. However, current pruning algorithms either only focus on one pruning category, e.g., structured pruning and unstructured, or need extensive hyperparameter tuning in order to get reasonable accuracy performance. To address these challenges, we propose LEArnable Pruning (LEAP), an effective method to gradually prune the model based on thresholds learned by gradient descent. Different than previous learnable pruning methods, which utilize $L_0$ or $L_1$ penalty to indirectly affect the final pruning ratio, LEAP introduces a novel regularization function, that directly interacts with the preset target pruning ratio. Moreover, in order to reduce hyperparameter tuning, a novel adaptive regularization coefficient is deployed to control the regularization penalty adaptively. With the new regularization term and its associated adaptive regularization coefficient, LEAP is able to be applied for different pruning granularity, including unstructured pruning, structured pruning, and hybrid pruning, with minimal hyperparameter tuning. We apply LEAP for BERT models on QQP/MNLI/SQuAD for different pruning settings. Our result shows that for all datasets, pruning granularity, and pruning ratios, LEAP achieves on-par or better results as compared to previous heavily hand-tuned methods.

📄 PDF Abstract BibTeX arXiv:2105.14636

Code (1)

yaozhewei/mlpruning 공식 구현 pytorch

Tasks

QQP

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Pruning 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Weight Decay 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

LEAP: Learnable End-to-End Adaptive Pruning of Large Language Models

2026-05-17 · Mohammad Mozaffari, Younes Hourri, Mohammad Rastegari, Mahyar Najibi arxiv

Unstructured sparsity is now natively accelerated by recent GPU kernels and dataflow hardware, shifting the bottleneck from inference execution to the pruning algorithm. State-of-the-art methods for unstructured LLM prun…

Learnable Permutation for Structured Sparsity on Transformer Models

2026-01-30 · Zekai Li, Ji Liu, Guanchen Li, Yixing Xu 외 arxiv

Structured sparsity has emerged as a popular model pruning technique, widely adopted in various architectures, including CNNs, Transformer models, and especially large language models (LLMs) in recent years. A promising …

LEAPS: End-to-End One-Step Person Search With Learnable Proposals

2023-03-21 · Zhiqiang Dong, Jiale Cao, Rao Muhammad Anwer, Jin Xie 외

We propose an end-to-end one-step person search approach with learnable proposals, named LEAPS. Given a set of sparse and learnable proposals, LEAPS employs a dynamic person search head to directly perform person detecti…

Human DetectionPerson Search

TinyFusion: Diffusion Transformers Learned Shallow

2024-12-02 · CVPR 2025 1 · Gongfan Fang, Kunjun Li, Xinyin Ma, Xinchao Wang

Diffusion Transformers have demonstrated remarkable capabilities in image generation but often come with excessive parameterization, resulting in considerable inference overhead in real-world applications. In this work, …

Image Generation

Leap: Inductive Link Prediction via Learnable TopologyAugmentation

2025-03-05 · Ahmed E. Samy, Zekarias T. Kefato, Sarunas Girdzijauskas

Link prediction is a crucial task in many downstream applications of graph machine learning. To this end, Graph Neural Network (GNN) is a widely used technique for link prediction, mainly in transductive settings, where …

Graph Neural NetworkInductive BiasInductive Link PredictionLink Prediction+1