GALOT: Generative Active Learning via Optimizable Zero-shot Text-to-image Generation

TL;DR

This paper introduces GALOT, a framework that combines zero-shot text-to-image synthesis with active learning to generate informative data samples from text descriptions, reducing data annotation costs and improving model training efficiency.

Contribution

The paper presents a novel end-to-end approach that leverages zero-shot T2I generation within active learning, enabling training solely from text descriptions.

Findings

Significant performance improvements over traditional active learning methods.

Efficient data generation reduces annotation costs.

Framework enables training from text descriptions alone.

Abstract

Active Learning (AL) represents a crucial methodology within machine learning, emphasizing the identification and utilization of the most informative samples for efficient model training. However, a significant challenge of AL is its dependence on the limited labeled data samples and data distribution, resulting in limited performance. To address this limitation, this paper integrates the zero-shot text-to-image (T2I) synthesis and active learning by designing a novel framework that can efficiently train a machine learning (ML) model sorely using the text description. Specifically, we leverage the AL criteria to optimize the text inputs for generating more informative and diverse data samples, annotated by the pseudo-label crafted from text, then served as a synthetic dataset for active learning. This approach reduces the cost of data collection and annotation while increasing the efficiency of model training by providing informative training samples, enabling a novel end-to-end ML task from text description to vision models. Through comprehensive evaluations, our framework demonstrates consistent and significant improvements over traditional AL methods.