# PLATON: Pruning Large Transformer Models with Upper Confidence Bound of Weight Importance

Source: [PLATON: Pruning Large Transformer Models with Upper Confidence Bound of Weight Importance](<https://www.sota2.com/research/paper/68fa6ce41e209131cea3690c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T00:21:36.840Z

Authors: Qingru Zhang, Simiao Zuo, Chen Liang, Alexander Bukharin, Pengcheng He, Weizhu Chen, Tuo Zhao

Published: 2022-06-25T05:38:39.000Z

## Abstract

Large Transformer-based models have exhibited superior performance in various natural language processing and computer vision tasks. However, these models contain enormous amounts of parameters, which restrict their deployment to real-world applications. To reduce the model size, researchers prune these models based on the weights' importance scores. However, such scores are usually estimated on mini-batches during training, which incurs large variability/uncertainty due to mini-batch sampling and complicated training dynamics. As a result, some crucial weights could be pruned by commonly used pruning methods because of such uncertainty, which makes training unstable and hurts generalization. To resolve this issue, we propose PLATON, which captures the uncertainty of importance scores by upper confidence bound (UCB) of importance estimation. In particular, for the weights with low importance scores but high uncertainty, PLATON tends to retain them and explores their capacity. We conduct extensive experiments with several Transformer-based models on natural language understanding, question answering and image classification to validate the effectiveness of PLATON. Results demonstrate that PLATON manifests notable improvement under different sparsity levels. Our code is publicly available at https://github.com/QingruZhang/PLATON.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Natural Language Understanding on GLUE](<https://www.sota2.com/research/sota/natural-language-understanding-on-glue.md>): Natural Language Understanding; GLUE; SST-2: 90.5.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2206.12562>)

[PDF](<https://arxiv.org/pdf/2206.12562.pdf>)
