LGDec 16, 2024

QPruner: Probabilistic Decision Quantization for Structured Pruning in Large Language Models

arXiv:2412.11629v112 citationsh-index: 3NAACL
Originality Incremental advance
AI Analysis

This work addresses memory efficiency challenges for deploying large language models, representing an incremental improvement in structured pruning techniques.

The paper tackles the problem of high memory consumption in structured pruning of large language models by introducing a quantization framework called QPruner, which uses layer-wise mixed-precision quantization and Bayesian optimization to achieve significant memory savings while maintaining or improving model performance on benchmark datasets.

The rise of large language models (LLMs) has significantly advanced various natural language processing (NLP) tasks. However, the resource demands of these models pose substantial challenges. Structured pruning is an effective approach to reducing model size, but it often results in significant accuracy degradation, necessitating parameter updates to adapt. Unfortunately, such fine-tuning requires substantial memory, which limits its applicability. To address these challenges, we introduce quantization into the structured pruning framework to reduce memory consumption during both fine-tuning and inference. However, the combined errors from pruning and quantization increase the difficulty of fine-tuning, requiring a more refined quantization scheme. To this end, we propose QPruner, a novel framework that employs structured pruning to reduce model size, followed by a layer-wise mixed-precision quantization scheme. Quantization precisions are assigned to each layer based on their importance to the target task, and Bayesian optimization is employed to refine precision allocation strategies, ensuring a balance between model accuracy and memory efficiency. Extensive experiments on benchmark datasets demonstrate that QPruner significantly outperforms existing methods in memory savings while maintaining or improving model performance.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes