Awesome-Novel-Class-Discovery

Novel Class Discovery (NCD) is a machine learning problem, where novel categories of instances are to be automatically discovered from an unlabelled pool. In contrast to clustering, NCD setting has access to labelled data from a disjoint set of classes. This topic has plausible real-world applications and is gathering much attention in the research community.

Here, we provide a non-exhaustive list of papers that study NCD.

Some Terms of Problem Setting

• Novel Class Discovery (NCD, aka Novel Category Discovery)
• Generalized Category Discovery (GCD, aka Generalized Class Discovery), Open-world Semi-supervised Learning (Open-word SSL)
• Novel Class Discovery without Forgetting (NCDwF), Class-incremental Novel CLass Discovery (Class-iNCD)
• Continuous Categories Discovery (CCD)
• Federated Generalized Category Discovery (Fed-GCD)
• Active Generalized Category Discovery (Active-GCD)
• TODO, such as Incremental Generalized Category Discovery (IGCD), Semantic Category Discovery (SCD)

Survey Papers

• Novel Class Discovery: an Introduction and Key Concepts [paper]
• Open-world Machine Learning: A Review and New Outlooks [paper]

Preprints

• HiLo: A Learning Framework for Generalized Category Discovery Robust to Domain Shifts [paper]
• Continual Novel Class Discovery via Feature Enhancement and Adaptation [paper]
• Exclusive Style Removal for Cross Domain Novel Class Discovery [paper]
• Revisiting Mutual Information Maximization for Generalized Category Discovery [paper]
• Beyond Known Clusters: Probe New Prototypes for Efficient Generalized Class Discovery [paper] [code]
• GET: Unlocking the Multi-modal Potential of CLIP for Generalized Category Discovery [paper] [code]
• Memory Consistency Guided Divide-and-Conquer Learning for Generalized Category Discovery [paper]
• YOLOOC: YOLO-based Open-Class Incremental Object Detection with Novel Class Discovery [paper]
• Beyond the Known: Novel Class Discovery for Open-world Graph Learning [paper]
• PANDAS: Prototype-based Novel Class Discovery and Detection [paper] [code]
• Learning from Semi-Factuals: A Debiased and Semantic-Aware Framework for Generalized Relation Discovery [paper]
• Federated Continual Novel Class Learning [paper]
• Generalized Category Discovery with Large Language Models in the Loop [paper]
• Towards Unbiased Training in Federated Open-world Semi-supervised Learning [paper]
• Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised Learning [paper]
• Novel class discovery meets foundation models for 3D semantic segmentation [paper]
• Generalized Category Discovery in Semantic Segmentation [paper] [code]
• Reinforcement Learning Based Multi-modal Feature Fusion Network for Novel Class Discovery [paper]
• Generalized Continual Category Discovery [paper]
• OpenGCD: Assisting Open World Recognition with Generalized Category Discovery [paper] [code]
• Novel Categories Discovery from probability matrix perspective [paper]
• CLIP-GCD: Simple Language Guided Generalized Category Discovery [paper]
• What's in a Name? Beyond Class Indices for Image Recognition [paper] (SCD, Semantic Category Discovery)
• NEV-NCD: Negative Learning, Entropy, and Variance regularization based novel action categories discovery [paper] [code]
• Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery [paper] [code]
• Automatically Discovering Novel Visual Categories with Self-supervised Prototype Learning [paper]
• CiPR: An Efficient Framework with Cross-instance Positive Relations for Generalized Category Discovery [paper]
• Textual Knowledge Matters: Cross-Modality Co-Teaching for Generalized Visual Class Discovery [paper]

2024

• Happy: A Debiased Learning Framework for Continual Generalized Category Discovery (NeurIPS 2024) [paper] [code]
• Online Continuous Generalized Category Discovery (ECCV 2024) [paper] [code]
• PromptCCD: Learning Gaussian Mixture Prompt Pool for Continual Category Discovery (ECCV 2024) [paper] [code]
• Self-Cooperation Knowledge Distillation for Novel Class Discovery (ECCV 2024) [paper]
• Dual-level Adaptive Self-Labeling for Novel Class Discovery in Point Cloud Segmentation (ECCV 2024) [paper] [code]
• Contextuality Helps Representation Learning for Generalized Category Discovery (ICIP 2024) [paper] [code]
• NC-NCD: Novel Class Discovery for Node Classification (CIKM 2024) [paper]
• A Practical Approach to Novel Class Discovery in Tabular Data (DMKD 2024) [paper] [code]
• Novel Class Discovery for Ultra-Fine-Grained Visual Categorization (CVPR 2024) [paper] [code]
• Contrastive Mean-Shift Learning for Generalized Category Discovery (CVPR 2024) [paper] [code]
• CDAD-Net: Bridging Domain Gaps in Generalized Category Discovery (CVPR Workshop 2024) [paper]
• Active Generalized Category Discovery (CVPR 2024) [paper] [code]
• Seeing Unseen: Discover Novel Biomedical Concepts via Geometry-Constrained Probabilistic Modeling (CVPR 2024) [paper]
• Federated Generalized Category Discovery (CVPR 2024) [paper]
• Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024) [paper] [project]
• SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuning (ICLR 2024) [paper] [code]
• A Unified Knowledge Transfer Network for Generalized Category Discovery (AAAI 2024)
• Novel Class Discovery in Chest X-Rays via Paired Images and Text (AAAI 2024) [framework]
• Semantic-Guided Novel Category Discovery (AAAI 2024) [paper] [code]
• Adaptive Discovering and Merging for Incremental Novel Class Discovery (AAAI 2024) [paper]
• Debiased Novel Category Discovering and Localization (AAAI 2024) [paper]
• Transfer and Alignment Network for Generalized Category Discovery (AAAI 2024) [paper] [code]
• Guided Cluster Aggregation: A Hierarchical Approach to Generalized Category Discovery (WACV 2024) [paper] [code]
• AMEND: Adaptive Margin and Expanded Neighborhood for Efficient Generalized Category Discovery (WACV 2024) [paper] [code]

2023

• Open-world Semi-supervised Generalized Relation Discovery Aligned in a Real-world Setting (EMNLP 2023) [paper] [code]
• A Graph-Theoretic Framework for Understanding Open-World Semi-Supervised Learning (NeurIPS 2023) [paper] [code]
• Decompose Novel into Known: Part Concept Learning For 3D Novel Class Discovery (NeurIPS 2023) [paper]
• Learn to Categorize or Categorize to Learn? Self-Coding for Generalized Category Discovery (NeurIPS 2023) [paper] [code]
• Towards Distribution-Agnostic Generalized Category Discovery (NeurIPS 2023) [paper] [code]
• No Representation Rules Them All in Category Discovery (NeurIPS 2023) [paper] [code]
• Discover and Align Taxonomic Context Priors for Open-world Semi-Supervised Learning (NeurIPS 2023) [paper] [code]
• Generalized Category Discovery with Clustering Assignment Consistency (ICONIP 2023) [paper]
• Towards Novel Class Discovery: A Study in Novel Skin Lesions Clustering (MICCAI 2023) [paper]
• Novel Class Discovery for Long-tailed Recognition (TMLR 2023) [paper]
• Generalized Categories Discovery for Long-tailed Recognition (ICCV Workshop 2023) [paper]
• Boosting Novel Category Discovery Over Domains with Soft Contrastive Learning and All-in-One Classifier (ICCV 2023) [paper] [code]
• Parametric Information Maximization for Generalized Category Discovery (ICCV 2023) [paper] [code]
• MetaGCD: Learning to Continually Learn in Generalized Category Discovery (ICCV 2023) [paper] [code]
• Proxy Anchor-based Unsupervised Learning for Continuous Generalized Category Discovery (ICCV 2023) [paper] [code]
• Class-relation Knowledge Distillation for Novel Class Discovery (ICCV 2023) [paper]
• Incremental Generalized Category Discovery (ICCV 2023) [paper] [code]
• Learning Semi-supervised Gaussian Mixture Models for Generalized Category Discovery (ICCV 2023) [paper] [code]
• Parametric Classification for Generalized Category Discovery: A Baseline Study (ICCV 2023) [paper] [code]
• An Interactive Interface for Novel Class Discovery in Tabular Data (ECML PKDD 2023, Demo Track) [paper] [code]
• When and How Does Known Class Help Discover Unknown Ones? Provable Understandings Through Spectral Analysis (ICML 2023) [paper] [code]
• Open-world Semi-supervised Novel Class Discovery (IJCAI 2023) [paper] [code]
• ImbaGCD: Imbalanced Generalized Category Discovery (CVPR Workshop 2023) [paper]
• On-the-Fly Category Discovery (CVPR 2023) [paper] [code]
• Bootstrap Your Own Prior: Towards Distribution-Agnostic Novel Class Discovery (CVPR 2023) [paper] [code]
• Dynamic Conceptional Contrastive Learning for Generalized Category Discovery (CVPR 2023) [paper] [code]
• PromptCAL: Contrastive Affinity Learning via Auxiliary Prompts for Generalized Novel Category Discovery (CVPR 2023) [paper] [code]
• Modeling Inter-Class and Intra-Class Constraints in Novel Class Discovery (CVPR 2023) [paper] [code]
• Novel Class Discovery for 3D Point Cloud Semantic Segmentation (CVPR 2023) [paper] [code]
• Generalized Category Discovery with Decoupled Prototypical Network (AAAI 2023) [paper] [code] (DPN)
• Supervised Knowledge May Hurt Novel Class Discovery Performance (TMLR 2023) [paper][code]
• OpenCon: Open-world Contrastive Learning (TMLR 2023) [paper] [code]

2022

• A Method for Discovering Novel Classes in Tabular Data (ICKG 2022) [paper] [code]
• Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning (EMNLP 2022) [paper]
• A Closer Look at Novel Class Discovery from the Labeled Set (NeurIPS Workshop 2022) [paper]
• Robust Semi-Supervised Learning when Not All Classes have Labels (NeurIPS 2022) [paper] [code]
• Grow and Merge: A Unified Framework for Continuous Categories Discovery (NeurIPS 2022) [paper] [code] (GM)
• XCon: Learning with Experts for Fine-grained Category Discovery (BMVC 2022) [paper] [code]
• Towards Realistic Semi-Supervised Learning (ECCV 2022) [paper] [code]
• Novel Class Discovery without Forgetting (ECCV 2022) [paper] (NCDwF)
• Class-incremental Novel Class Discovery (ECCV 2022) [paper] [code] (FRoST)
• OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning (ECCV 2022) [paper] [code]
• Residual Tuning: Toward Novel Category Discovery Without Labels (TNNLS 2022) [paper] [code] (ResTune)
• Open-World Semi-Supervised Learning (ICLR 2022) [paper] [code]
• Meta Discovery: Learning to Discover Novel Classes given Very Limited Data (ICLR 2022) [paper] [code] (MEDI)
• Self-Labeling Framework for Novel Category Discovery over Domains (AAAI 2022) [paper]
• Towards Open-Set Object Detection and Discovery (CVPR Workshop 2022) [paper]
• Divide and Conquer: Compositional Experts for Generalized Novel Class Discovery (CVPR 2022) [paper] [code] (ComEx)
• Novel Class Discovery in Semantic Segmentation (CVPR 2022) [paper] [code]
• Generalized Category Discovery (CVPR 2022) [paper] [code] (GCD)
• Spacing Loss for Discovering Novel Categories (CVPR Workshop 2022) [paper] (Spacing Loss)
• Open Set Domain Adaptation By Novel Class Discovery (ICME 2022) [paper]
• Progressive Self-Supervised Clustering With Novel Category Discovery (TCYB 2022) [paper] [code]
• Novel Class Discovery: A Dependency Approach (ICASSP 2022) [paper]

2021

• Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillation (NeurIPS 2021) [paper] [code] (DualRS)
• A Unified Objective for Novel Class Discovery (ICCV 2021) [paper] [code] (UNO)
• Joint Representation Learning and Novel Category Discovery on Single- and Multi-modal Data (ICCV 2021) [paper] (Joint)
• Neighborhood Contrastive Learning for Novel Class Discovery (CVPR 2021) [paper] [code] (NCL)
• OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in An Open World (CVPR 2021) [paper] (OpenMix)
• AutoNovel: Automatically Discovering and Learning Novel Visual Categories (TPAMI 2021) [paper] (AutoNovel aka RS)
• End-to-end novel visual categories learning via auxiliary self-supervision (Neural Networks 2021) [paper]

2020

• Automatically Discovering and Learning New Visual Categories with Ranking Statistics (ICLR 2020) [paper] [code] (AutoNovel aka RS)
• Open-World Class Discovery with Kernel Networks (ICDM 2020) [paper] [code]

2019

• Learning to discover novel visual categories via deep transfer clustering (ICCV 2019) [paper] [code] (DTC)
• Multi-class classification without multi-class labels (ICLR 2019) [paper] [code] (MCL)

2018

• Learning to cluster in order to transfer across domains and tasks (ICLR 2018) [paper] [code] (KCL)

2016

• Neural network-based clustering using pairwise constraints (ICLR-workshop 2016) [paper] [code]

Contributing

Please help us improve the above listing by submitting PRs of other papers in this space. Thank you!