让模型在训练时就学会发现新类别，无需哈希编码，效果更优。
Learning through Creation: A Hash-Free Framework for On-the-Fly Category Discovery
- 训练时动态生成伪新类样本，直接提升模型发现能力。
- 在7个基准上准确率提升1.5%到13.1%，显著优于现有方法。
- 适合需要实时识别未知类别的实际应用场景。
在线即时类别发现（OCD）旨在推理过程中识别已知类别并同时发现新出现的未知类别，仅依赖训练阶段已知类别的监督信号。现有方法要么依赖固定标签，要么使用基于扩散的数据增强，但均未在训练中显式优化发现任务。这种离线学习与在线发现之间的目标错配，导致模型难以胜任发现任务。此外，多数方法依赖哈希编码或严重特征压缩，限制了表示能力。为此，我们提出学习通过创造（LTC），一种全特征、无哈希的框架，将新类别意识注入离线训练过程。核心是一个轻量级在线伪未知生成器，基于核能最小化与熵最大化（MKEE）驱动，随模型动态演化，在推理时以极低成本实时生成伪新类样本。这些样本通过自适应阈值的双最大间隔目标融合，强化模型对未知区域的区分与检测能力。在七个基准上的实验表明，LTC始终优于先前方法，所有类别准确率提升1.5%至13.1%。代码已公开于https://github.com/brandinzhang/LTC。
原文摘要 · Abstract (English)
On-the-Fly Category Discovery (OCD) aims to recognize known classes while simultaneously discovering emerging novel categories during inference, using supervision only from known classes during offline training. Existing approaches rely either on fixed label supervision or on diffusion-based augmentations to enhance the backbone, yet none of them explicitly train the model to perform the discovery task required at test time. It is fundamentally unreasonable to expect a model optimized on limited labeled data to carry out a qualitatively different discovery objective during inference. This mismatch creates a clear optimization misalignment between the offline learning stage and the online discovery stage. In addition, prior methods often depend on hash-based encodings or severe feature compression, which further limits representational capacity. To address these issues, we propose Learning through Creation (LTC), a fully feature-based and hash-free framework that injects novel-category awareness directly into offline learning. At its core is a lightweight, online pseudo-unknown generator driven by kernel-energy minimization and entropy maximization (MKEE). Unlike previous methods that generate synthetic samples once before training, our generator evolves jointly with the model dynamics and synthesizes pseudo-novel instances on the fly at negligible cost. These samples are incorporated through a dual max-margin objective with adaptive thresholding, strengthening the model's ability to delineate and detect unknown regions through explicit creation. Extensive experiments across seven benchmarks show that LTC consistently outperforms prior work, achieving improvements ranging from 1.5 percent to 13.1 percent in all-class accuracy. The code is available at https://github.com/brandinzhang/LTC
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。