arXiv：cs.AI· Tianle Liu, Fan Lyu, Tingliang Feng, Chenggong Ni, Tianqi Wang, Zhang Zhang, Fuyuan Hu, Rui Yao, Tao Zhou·· 16 小时前
Energy Ring：稳定特征能量，实现可靠的测试时新类发现
Energy Ring: Stabilizing Feature Energy for Reliable Test-Time Discovery
AI 导读
针对测试时新类发现（TTD）中已知类与未知类特征易坍塌混叠的问题，研究者提出 Energy Ring，将两类样本锚定在分离的能量环上以稳定能量间隔。该方法用轻量 adapter 扩大范数差距并保留已知类方向，配合扰动一致性模块为未知类形成紧凑聚类，测试阶段则以能量引导置信度加权和稳定性校正机制提升预测。实验显示其 TTD 性能显著优于现有方法。
正文
Abstract:Test-Time Discovery (TTD) aims to recognize known classes and detect novel ones while the model is already deployed, operating directly on a continual unlabeled test-time stream. Existing TTD methods are overly conservative, making known and unknown features collapse together and difficult to separate during test-time adaptation. In this paper, we revisit TTD from an energy perspective and find that the natural energy gap between known and unknown classes is unstable and easily collapses. We therefore propose Energy Ring to stabilize this gap by anchoring the two types of classes on separate energy rings. A lightweight adapter enlarges the norm-based gap while preserving known-class directions, and a perturbation consistency module forms compact clusters for unknown classes. During testing, energy-guided confidence weighting improves prediction, and a stability-based rectification mechanism corrects unreliable updates. Experimental results show that Energy Ring significantly improves TTD performance against existing TTD methods.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.14525 [cs.AI] |
| (or arXiv:2610.14525v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.14525 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tianle Liu [view email]
[v1]
Sat, 10 Oct 2026 03:36:12 UTC (2,221 KB)
来源：arXiv：cs.AI · arxiv.org