# Generalized Category Discovery with Large Language Models in the Loop

arXiv: https://arxiv.org/abs/2312.10897 · HTML: https://mlsanity.com/paper/2312.10897/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 42 references that are in the map.
- First posted on arXiv 18 Dec 2023.

## Builds on (most cited first, 22 of 22)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [Deep Clustering for Unsupervised Learning of Visual Features](https://mlsanity.com/paper/1807.05520/) (2018)
- [Unsupervised Deep Embedding for Clustering Analysis](https://mlsanity.com/paper/1511.06335/) (2015)
- [Efficient Intent Detection with Dual Sentence Encoders](https://mlsanity.com/paper/2003.04807/) (2020)
- [An Evaluation Dataset for Intent Classification and Out-of-Scope Prediction](https://mlsanity.com/paper/1909.02027/) (2019)
- [ChatIE: Zero-Shot Information Extraction via Chatting with ChatGPT](https://mlsanity.com/paper/2302.10205/) (2023)
- [Towards K-means-friendly Spaces: Simultaneous Deep Learning and Clustering](https://mlsanity.com/paper/1610.04794/) (2016)
- [Learning to Discover Novel Visual Categories via Deep Transfer Clustering](https://mlsanity.com/paper/1908.09884/) (2019)
- [Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification](https://mlsanity.com/paper/2001.01526/) (2020)
- [Open-World Semi-Supervised Learning](https://mlsanity.com/paper/2102.03526/) (2021)
- [Active Learning by Acquiring Contrastive Examples](https://mlsanity.com/paper/2109.03764/) (2021)
- [Neighborhood Contrastive Learning for Novel Class Discovery](https://mlsanity.com/paper/2106.10731/) (2021)
- [OVANet: One-vs-All Network for Universal Domain Adaptation](https://mlsanity.com/paper/2104.03344/) (2021)
- [ClusterLLM: Large Language Models as a Guide for Text Clustering](https://mlsanity.com/paper/2305.14871/) (2023)
- [Discovering New Intents with Deep Aligned Clustering](https://mlsanity.com/paper/2012.08987/) (2020)
- [Discovering New Intents via Constrained Deep Adaptive Clustering with Cluster Refinement](https://mlsanity.com/paper/1911.08891/) (2019)
- [Generalized Category Discovery with Decoupled Prototypical Network](https://mlsanity.com/paper/2211.15115/) (2022)
- [Improving Contrastive Learning of Sentence Embeddings from AI Feedback](https://mlsanity.com/paper/2305.01918/) (2023)
- [Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning](https://mlsanity.com/paper/2210.07733/) (2022)
- [A Diffusion Weighted Graph Framework for New Intent Discovery](https://mlsanity.com/paper/2310.15836/) (2023)
- [DNA: Denoised Neighborhood Aggregation for Fine-grained Category Discovery](https://mlsanity.com/paper/2310.10151/) (2023)

## Built on by (most cited first, 15 of 15)

- [Are Large Language Models Good Prompt Optimizers?](https://mlsanity.com/paper/2402.02101/) (2024)
- [SelEx: Self-Expertise in Fine-Grained Generalized Category Discovery](https://mlsanity.com/paper/2408.14371/) (2024)
- [Flipped Classroom: Aligning Teacher Attention with Student in Generalized Category Discovery](https://mlsanity.com/paper/2409.19659/) (2024)
- [A Review of Multimodal Explainable Artificial Intelligence: Past, Present and Future](https://mlsanity.com/paper/2412.14056/) (2024)
- [A Survey of Ontology Expansion for Conversational Understanding](https://mlsanity.com/paper/2410.15019/) (2024)
- [GLEAN: Active Generalized Category Discovery with Diverse LLM Feedback](https://mlsanity.com/paper/2502.18414/) (2025)
- [Generative Induction of Dialogue Task Schemas with Streaming Refinement and Simulated Interactions](https://mlsanity.com/paper/2504.18474/) (2025)
- [Empowering Multimodal LLMs with External Tools: A Comprehensive Survey](https://mlsanity.com/paper/2508.10955/) (2025)
- [LLMs Enable Bag-of-Texts Representations for Short-Text Clustering](https://mlsanity.com/paper/2510.06747/) (2025)
- [Pseudo-Label Enhanced Prototypical Contrastive Learning for Uniformed Intent Discovery](https://mlsanity.com/paper/2410.20219/) (2024)
- [Generalized Category Discovery in Event-Centric Contexts: Latent Pattern Mining with LLMs](https://mlsanity.com/paper/2505.23304/) (2025)
- [Towards Automated Error Discovery: A Study in Conversational AI](https://mlsanity.com/paper/2509.10833/) (2025)
- [ReLKD: Inter-Class Relation Learning with Knowledge Distillation for Generalized Category Discovery](https://mlsanity.com/paper/2512.07229/) (2025)
- [OmniGCD: Abstracting Generalized Category Discovery for Modality Agnosticism](https://mlsanity.com/paper/2604.14762/) (2026)
- [GRACE: LLM-Grounded Semantic Metric Spaces for Scalable Mixed-Data Clustering](https://mlsanity.com/paper/2608.07881/) (2026)

## Abstract (arXiv)

Generalized Category Discovery (GCD) is a crucial task that aims to recognize both known and novel categories from a set of unlabeled data by utilizing a few labeled data with only known categories. Due to the lack of supervision and category information, current methods usually perform poorly on novel categories and struggle to reveal semantic meanings of the discovered clusters, which limits their applications in the real world. To mitigate the above issues, we propose Loop, an end-to-end active-learning framework that introduces Large Language Models (LLMs) into the training loop, which can boost model performance and generate category names without relying on any human efforts. Specifically, we first propose Local Inconsistent Sampling (LIS) to select samples that have a higher probability of falling to wrong clusters, based on neighborhood prediction consistency and entropy of cluster assignment probabilities. Then we propose a Scalable Query strategy to allow LLMs to choose true neighbors of the selected samples from multiple candidate samples. Based on the feedback from LLMs, we perform Refined Neighborhood Contrastive Learning (RNCL) to pull samples and their neighbors closer to learn clustering-friendly representations. Finally, we select representative samples from clusters corresponding to novel categories to allow LLMs to generate category names for them. Extensive experiments on three benchmark datasets show that Loop outperforms SOTA models by a large margin and generates accurate category names for the discovered clusters. Code and data are available at https://github.com/Lackel/LOOP.