# How many classes do we need to see for novel class discovery?

arXiv: https://arxiv.org/abs/2509.15585 · HTML: https://mlsanity.com/paper/2509.15585/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 22 references that are in the map.
- First posted on arXiv 19 Sep 2025.

## Builds on (most cited first, 16 of 16)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [Learning to Discover Novel Visual Categories via Deep Transfer Clustering](https://mlsanity.com/paper/1908.09884/) (2019)
- [Generalized Category Discovery](https://mlsanity.com/paper/2201.02609/) (2022)
- [A Unified Objective for Novel Class Discovery](https://mlsanity.com/paper/2108.08536/) (2021)
- [Learning to cluster in order to transfer across domains and tasks](https://mlsanity.com/paper/1711.10125/) (2017)
- [AutoNovel: Automatically Discovering and Learning Novel Visual Categories](https://mlsanity.com/paper/2106.15252/) (2021)
- [Towards Visually Explaining Variational Autoencoders](https://mlsanity.com/paper/1911.07389/) (2019)
- [Multi-class Classification without Multi-class Labels](https://mlsanity.com/paper/1901.00544/) (2019)
- [Novel Class Discovery: an Introduction and Key Concepts](https://mlsanity.com/paper/2302.12028/) (2023)
- [Open-world Semi-supervised Novel Class Discovery](https://mlsanity.com/paper/2305.13095/) (2023)
- [Failing Conceptually: Concept-Based Explanations of Dataset Shift](https://mlsanity.com/paper/2104.08952/) (2021)
- [Open-Source Tools for Behavioral Video Analysis: Setup, Methods, and Development](https://mlsanity.com/paper/2204.02842/) (2022)
- [OpenGCD: Assisting Open World Recognition with Generalized Category Discovery](https://mlsanity.com/paper/2308.06926/) (2023)
- [A Practical Approach to Novel Class Discovery in Tabular Data](https://mlsanity.com/paper/2311.05440/) (2023)

## Abstract (arXiv)

Novel class discovery is essential for ML models to adapt to evolving real-world data, with applications ranging from scientific discovery to robotics. However, these datasets contain complex and entangled factors of variation, making a systematic study of class discovery difficult. As a result, many fundamental questions are yet to be answered on why and when new class discoveries are more likely to be successful. To address this, we propose a simple controlled experimental framework using the dSprites dataset with procedurally generated modifying factors. This allows us to investigate what influences successful class discovery. In particular, we study the relationship between the number of known/unknown classes and discovery performance, as well as the impact of known class 'coverage' on discovering new classes. Our empirical results indicate that the benefit of the number of known classes reaches a saturation point beyond which discovery performance plateaus. The pattern of diminishing return across different settings provides an insight for cost-benefit analysis for practitioners and a starting point for more rigorous future research of class discovery on complex real-world datasets.