paper-with-me

Papers

All-in-One Tuning and Structural Pruning for Domain-Specific LLMs

2024-12-19 · Lei Lu, Zhepeng Wang, Runxue Bao, Mengbing Wang, Fangyi Li, Yawen Wu, Weiwen Jiang, Jie Xu, Yanzhi Wang, Shangqian Gao

Existing pruning techniques for large language models (LLMs) targeting domain-specific applications typically follow a two-stage process: pruning the pretrained general-purpose LLMs and then fine-tuning the pruned LLMs on specific domains. However, the pruning decisions, derived from the pretrained weights, remain unchanged during fine-tuning, even if the weights have been updated. Therefore, such a combination of the pruning decisions and the finetuned weights may be suboptimal, leading to non-negligible performance degradation. To address these limitations, we propose ATP: All-in-One Tuning and Structural Pruning, a unified one-stage structural pruning and fine-tuning approach that dynamically identifies the current optimal substructure throughout the fine-tuning phase via a trainable pruning decision generator. Moreover, given the limited available data for domain-specific applications, Low-Rank Adaptation (LoRA) becomes a common technique to fine-tune the LLMs. In ATP, we introduce LoRA-aware forward and sparsity regularization to ensure that the substructures corresponding to the learned pruning decisions can be directly removed after the ATP process. ATP outperforms the state-of-the-art two-stage pruning methods on tasks in the legal and healthcare domains. More specifically, ATP recovers up to 88% and 91% performance of the dense model when pruning 40% parameters of LLaMA2-7B and LLaMA3-8B models, respectively.

📄 PDF Abstract BibTeX arXiv:2412.14426

Code (0)

등록된 구현이 없습니다.

Tasks

All

Methods 이 논문이 사용한 방법론

Pruning 설명 없음

Similar Papers 제목 키워드 기반

Investigating Structural Pruning and Recovery Techniques for Compressing Multimodal Large Language Models: An Empirical Study

2025-07-28 · Yiran Huang, Lukas Thede, Massimiliano Mancini, Wenjia Xu 외 arxiv

While Multimodal Large Language Models (MLLMs) demonstrate impressive capabilities, their substantial computational and memory requirements pose significant barriers to practical deployment. Current parameter reduction t…

Knowledge Distillation

PAT: Pruning-Aware Tuning for Large Language Models

2024-08-27 · Yijiang Liu, Huanrui Yang, Youxin Chen, Rongyu Zhang 외

Large language models (LLMs) excel in language tasks, especially with supervised fine-tuning after pre-training. However, their substantial memory and computational requirements hinder practical applications. Structural …

RankAdaptor: Hierarchical Rank Allocation for Efficient Fine-Tuning Pruned LLMs via Performance Model

2024-06-22 · Changhai Zhou, Shijie Han, Lining Yang, Yuhua Zhou 외

The efficient compression of large language models (LLMs) has become increasingly popular. However, recovering the performance of compressed LLMs remains a major challenge. The current practice in LLM compression entails…

Incremental LearningMeta-LearningScheduling

LoRAPrune: Structured Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning

2023-05-28 · Mingyang Zhang, Hao Chen, Chunhua Shen, Zhen Yang 외

Large Language Models (LLMs), such as LLaMA and T5, have shown exceptional performance across various tasks through fine-tuning. Although low-rank adaption (LoRA) has emerged to cheaply fine-tune these LLMs on downstream…

Model CompressionNetwork Pruningparameter-efficient fine-tuning

Detecting and Pruning Prominent but Detrimental Neurons in Large Language Models

2025-07-12 · Ameen Ali, Shahar Katz, Lior Wolf, Ivan Titov arxiv

Large language models (LLMs) often develop learned mechanisms specialized to specific datasets, such as reliance on domain-specific correlations, which yield high-confidence predictions without generalizable reasoning. W…