Learning Like Humans: Analogical Concept Learning for Generalized Category Discovery
This work addresses the challenge of discovering novel categories in unlabeled data while maintaining known categories, particularly for fine-grained, look-alike categories, offering a plug-and-play solution for GCD pipelines.
The paper tackles the problem of Generalized Category Discovery (GCD) by introducing the Analogical Textual Concept Generator (ATCG), which analogizes from labeled knowledge to generate textual concepts for unlabeled samples, improving performance across six benchmarks with the largest gains on fine-grained data.
Generalized Category Discovery (GCD) seeks to uncover novel categories in unlabeled data while preserving recognition of known categories, yet prevailing visual-only pipelines and the loose coupling between supervised learning and discovery often yield brittle boundaries on fine-grained, look-alike categories. We introduce the Analogical Textual Concept Generator (ATCG), a plug-and-play module that analogizes from labeled knowledge to new observations, forming textual concepts for unlabeled samples. Fusing these analogical textual concepts with visual features turns discovery into a visual-textual reasoning process, transferring prior knowledge to novel data and sharpening category separation. ATCG attaches to both parametric and clustering style GCD pipelines and requires no changes to their overall design. Across six benchmarks, ATCG consistently improves overall, known-class, and novel-class performance, with the largest gains on fine-grained data. Our code is available at: https://github.com/zhou-9527/AnaLogical-GCD.