GLEAN: Active Generalized Category Discovery with Diverse LLM Feedback

TL;DR

GLEAN is a unified framework that actively utilizes diverse large language model feedback to improve generalized category discovery, effectively recognizing known and novel categories with minimal supervision.

Contribution

It introduces a novel active learning approach leveraging multiple types of LLM feedback to enhance GCD performance without extensive human annotations.

Findings

GLEAN outperforms state-of-the-art models across various datasets.

It effectively uses LLM feedback to improve cluster semantic understanding.

GLEAN reduces the need for costly human annotations in GCD tasks.

Abstract

Generalized Category Discovery (GCD) is a practical and challenging open-world task that aims to recognize both known and novel categories in unlabeled data using limited labeled data from known categories. Due to the lack of supervision, previous GCD methods face significant challenges, such as difficulty in rectifying errors for confusing instances, and inability to effectively uncover and leverage the semantic meanings of discovered clusters. Therefore, additional annotations are usually required for real-world applicability. However, human annotation is extremely costly and inefficient. To address these issues, we propose GLEAN, a unified framework for generalized category discovery that actively learns from diverse and collaborative LLM feedback. Our approach leverages three different types of LLM feedback to: (1) improve instance-level contrastive features, (2) generate category descriptions, and (3) align uncertain instances with LLM-selected category descriptions. Extensive experiments demonstrate the superior performance of GLEAN over state-of-the-art models across diverse datasets, metrics, and supervision settings. Our code is available at https://github.com/amazon-science/Glean.