Aug 24, 2024 · In this work, we introduce Online Continuous Generalized Category Discovery (OCGCD), which considers the dynamic nature of data streams where ...

Aug 23, 2023 · Generalized Continual Category Discovery (GCCD) tackles learning from sequentially arriving, partially labeled datasets while uncovering new ...

The goal is to continually discover novel classes while maintaining the performance in known classes. We name the setting Continual Generalized Category Discov-.

The energy-guided discovery splits unlabeled data into known, seen, and unseen data for better novel category discovery, while variance-based feature ...

Nov 1, 2024 · In this work, we introduce Online Continuous Generalized Category Discovery (OCGCD), which considers the dynamic nature of data streams where ...

The goal is to continually discover novel classes while maintaining the performance in known classes. We name the setting Continual Generalized Category ...

Nov 5, 2024 · This paper studies a new task called Continual Generalized Category Discovery (C-GCD), which aims to incrementally discover more new classes ...

We call this setting Generalized Continual Category Discovery (GCCD). It unifies CL and GCD, bridging the gap between synthetic benchmarks and real-life ...

We explore the problem of Incremental Generalized Cat- egory Discovery (IGCD). This is a challenging category- incremental learning setting where the goal is to ...

This repository is the official implementation of MetaGCD: Learning to Continually Learn in Generalized Category Discovery.