Learning with Less: Knowledge Distillation from Large Language Models via Unlabeled Data

TL;DR

This paper introduces LLKD, a method for efficient knowledge distillation from large language models to smaller models using unlabeled data, focusing on adaptive sample selection to improve data efficiency and model performance.

Contribution

The paper proposes LLKD, an adaptive sample selection technique that enhances knowledge distillation from LLMs by prioritizing high-confidence and informative samples, reducing data and computational requirements.

Findings

LLKD outperforms baseline methods in multiple NLP tasks.

It achieves higher data efficiency with fewer labeled samples.

The method improves smaller model performance while reducing resource usage.

Abstract

In real-world NLP applications, Large Language Models (LLMs) offer promising solutions due to their extensive training on vast datasets. However, the large size and high computation demands of LLMs limit their practicality in many applications, especially when further fine-tuning is required. To address these limitations, smaller models are typically preferred for deployment. However, their training is hindered by the scarcity of labeled data. In contrast, unlabeled data is often readily which can be leveraged by using LLMs to generate pseudo-labels for training smaller models. This enables the smaller models (student) to acquire knowledge from LLMs(teacher) while reducing computational costs. This process introduces challenges, such as potential noisy pseudo-labels. Selecting high-quality and informative data is therefore critical to enhance model performance while improving the efficiency of data utilization. To address this, we propose LLKD that enables Learning with Less computational resources and less data for Knowledge Distillation from LLMs. LLKD is an adaptive sample selection method that incorporates signals from both the teacher and student. Specifically, it prioritizes samples where the teacher demonstrates high confidence in its labeling, indicating reliable labels, and where the student exhibits a high information need, identifying challenging samples that require further learning. Our comprehensive experiments show that LLKD achieves superior performance across various datasets with higher data efficiency.