arXiv cs.AI· Tianle Liu, Gongbo Cheng, Chenggong Ni, Minming Gu, Tao Zhou, Fuyuan Hu, Fan Lyu·· 3 天前AI 评分22
SEGA：用关系语义锚定实现测试时发现
Relational Semantic Grounding for Test-Time Discovery
AI 导读
研究者提出 SEGA 框架，利用大语言模型为已知类生成属性丰富的描述，并在训练中构建属性锚定的视觉-语义空间。测试时通过 Relational Semantic Grounding（RSG）用相似与不相似的已知类锚点定位未知样本，让 LLM 生成候选概念并在语义空间中重排序，再以置信度引导做测试时语义稳定化。该方法将匿名测试时聚类转化为可解释的语义概念，在 TTD 基线上取得提升。
正文 · 原文
Abstract:Test-Time Discovery (TTD) aims to recognize known classes and detect novel ones from a continuous unlabeled test stream after deployment. Existing TTD methods mainly rely on visual clustering or prototype comparison, and can only assign anonymous labels to discovered classes, leaving them difficult to interpret in open-world applications. In this paper, we propose SEGA, a Semantic Grounding with Attributes framework for TTD. SEGA leverages large language models to generate attribute-rich descriptions for known classes and constructs an attribute-anchored visual-semantic space during training. At test time, we introduce Relational Semantic Grounding (RSG), which grounds unknown samples with similar and dissimilar known-class anchors, prompts the LLM to generate candidate concepts, and reranks them in the learned semantic space. The selected concept is further used for confidence-guided test-time semantic stabilization. By coupling attribute-level semantic alignment with relational concept grounding, SEGA transforms anonymous test-time clusters into interpretable semantic concepts. Experiments show that SEGA improves TTD baselines with the promise of semantic grounding for open-world.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.14721 [cs.AI] |
| (or arXiv:2610.14721v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.14721 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tianle Liu [view email]
[v1]
Sat, 10 Oct 2026 08:42:16 UTC (2,196 KB)
来源：arXiv cs.AI · arxiv.org