Projected Compression: Trainable Projection for Efficient Transformer Compression

• URL: http://arxiv.org/abs/2506.22255v1
• Date: Fri, 27 Jun 2025 14:24:01 GMT
• Title: Projected Compression: Trainable Projection for Efficient Transformer Compression
• Authors: Maciej Stefaniak, Michał Krutul, Jan Małaśnicki, Maciej Pióro, Jakub Krajewski, Sebastian Jaszczur, Marek Cygan, Kamil Adamczewski, Jan Ludziejewski,
• Abstract summary: Large language models have steadily increased in size to achieve improved performance.<n>Projected Compression is a novel model compression technique that reduces model weights by utilizing projection modules.<n> Experimental results show that Projected Compression outperforms the comparable hard pruning and retraining approach on higher quality models.
• Score: 2.9812951075697325
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models have steadily increased in size to achieve improved performance; however, this growth has also led to greater inference time and computational demands. Consequently, there is rising interest in model size reduction methods. To address this issue, we propose Projected Compression, a novel model compression technique, that reduces model weights by utilizing projection modules. Specifically, we first train additional trainable projections weights and preserve access to all the original model parameters. Subsequently, these projections are merged into a lower-dimensional product matrix, resulting in a reduced-size standard Transformer-based model. Unlike alternative approaches that require additional computational overhead, our method matches the base model's per-token computation step in FLOPs. Experimental results show that Projected Compression outperforms the comparable hard pruning and retraining approach on higher quality models. Moreover, the performance margin scales well with the number of tokens.

Related papers

• On Information Geometry and Iterative Optimization in Model Compression: Operator Factorization [5.952537659103525]
  We argue that many successful model compression approaches can be understood as implicitly approximating information divergences for this projection.<n>We prove convergence of iterative singular value thresholding for training neural networks subject to a soft rank constraint.
  arXiv  Detail & Related papers  (2025-07-12T23:39:14Z)
• TuneComp: Joint Fine-tuning and Compression for Large Foundation Models [50.33925662486034]
  sequential fine-tuning and compression sacrifices performance, while creating a larger than necessary model as an intermediate step.<n>We propose to jointly fine-tune and compress the model by gradually distilling it to a pruned low-rank structure.<n> Experiments demonstrate that joint fine-tuning and compression significantly outperforms other sequential compression methods.
  arXiv  Detail & Related papers  (2025-05-27T23:49:35Z)
• Dynamic Base model Shift for Delta Compression [53.505380509713575]
  Delta compression attempts to lower the costs by reducing the redundancy of delta parameters.<n>Existing methods by default employ the pretrained model as the base model and compress the delta parameters for every task.<n>We propose Dynamic Base Model Shift (DBMS), which dynamically adapts the base model to the target task before performing delta compression.
  arXiv  Detail & Related papers  (2025-05-16T15:11:19Z)
• Choose Your Model Size: Any Compression by a Single Gradient Descent [9.074689052563878]
  We present Any Compression via Iterative Pruning (ACIP)<n>ACIP is an algorithmic approach to determine a compression-performance trade-off from a single gradient descent run.<n>We show that ACIP seamlessly complements common quantization-based compression techniques.
  arXiv  Detail & Related papers  (2025-02-03T18:40:58Z)
• Merging Feed-Forward Sublayers for Compressed Transformers [16.746335565636976]
  We present a novel approach to model compression by merging similar parameter groups within a model.<n>Specifically, we select, align, and merge separate feed-forward sublayers in Transformer models.<n>We demonstrate performance comparable to the original models while combining more than a third of model feed-forward sublayers.
  arXiv  Detail & Related papers  (2025-01-10T17:25:11Z)
• Singular Value Scaling: Efficient Generative Model Compression via Pruned Weights Refinement [9.454314879815337]
  generative models often exhibit dominant singular vectors, hindering fine-tuning efficiency and leading to suboptimal performance.<n>We introduce Singular Value Scaling (SVS), a versatile technique for refining pruned weights, applicable to both model types.<n>SVS improves compression performance across model types without additional training costs.
  arXiv  Detail & Related papers  (2024-12-23T08:40:08Z)
• Diffusion Product Quantization [18.32568431229839]
  We explore the quantization of diffusion models in extreme compression regimes to reduce model size while maintaining performance. We apply our compression method to the DiT model on ImageNet and consistently outperform other quantization approaches.
  arXiv  Detail & Related papers  (2024-11-19T07:47:37Z)
• Scaling Pre-trained Language Models to Deeper via Parameter-efficient Architecture [68.13678918660872]
  We design a more capable parameter-sharing architecture based on matrix product operator (MPO) MPO decomposition can reorganize and factorize the information of a parameter matrix into two parts. Our architecture shares the central tensor across all layers for reducing the model size.
  arXiv  Detail & Related papers  (2023-03-27T02:34:09Z)
• Language model compression with weighted low-rank factorization [73.61874728240568]
  We introduce Fisher information to weigh the importance of parameters affecting the model prediction. We find that our resulting task accuracy is much closer to the original model's performance. Our method can directly compress a task-specific model while achieving better performance than other compact model strategies.
  arXiv  Detail & Related papers  (2022-06-30T21:57:07Z)
• Dynamic Model Pruning with Feedback [64.019079257231]
  We propose a novel model compression method that generates a sparse trained model without additional overhead. We evaluate our method on CIFAR-10 and ImageNet, and show that the obtained sparse models can reach the state-of-the-art performance of dense models.
  arXiv  Detail & Related papers  (2020-06-12T15:07:08Z)
• Train Large, Then Compress: Rethinking Model Size for Efficient Training and Inference of Transformers [94.43313684188819]
  We study the impact of model size in this setting, focusing on Transformer models for NLP tasks that are limited by compute. We first show that even though smaller Transformer models execute faster per iteration, wider and deeper models converge in significantly fewer steps. This leads to an apparent trade-off between the training efficiency of large Transformer models and the inference efficiency of small Transformer models.
  arXiv  Detail & Related papers  (2020-02-26T21:17:13Z)

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.