Tag
提出基于局部本征维数（LID）的无监督特征选择方法LIDFS，通过对特征点进行最大似然LID估计排序特征，在23个高维数据集上取得领先的分类与聚类性能，且运行速度快于最接近的竞品方法。
A Reddit post shares the UnEZ paper, a fully unsupervised and directly interpretable neural network similar to an autoencoder but with an interpretable latent space, and asks the community about related architectures.
该论文研究在769维特征下，完整人类基因评分数据集与激酶基因子集是否呈现相同的聚类结构，通过四种聚类方法比较发现最优聚类数均为两类。
该论文提出将关联规则挖掘(APriori/FP-Growth)作为无需标注的无监督框架,用于新闻框架检测,并在加沙战争报道的2.5万余篇多语种新闻上验证了其识别跨媒体军事框架与事件驱动框架转变的能力。
AnoMix 提出一种多教师排序蒸馏方法（CoRank），将多个异构无监督时间序列异常检测器蒸馏为单一轻量学生模型（<80k 参数），在七个基准数据集上优于单个检测器和静态分数集成，其零样本变体还能在未见数据集上超越大型时间序列基础模型。
提出了TEAL框架，通过Explorer与Critic两个代理在无标签条件下为数据湖分析的编码代理自动构建和优化可复用工具库，并引入TEAL-Q生成合成问题，在KramaBench和LakeQA上显著提升多个Worker LLM在紧凑步数预算下的表现。
提出一种无需人工偏好标注的多模态奖励建模方法：在冻结的 Qwen3-VL 骨干网络的中间层上训练线性探针，通过扰动字幕对比进行训练，在多个基准上达到或超过 GPT-4o，且微调计算量仅为千分之一。
该论文基准测试了超双曲空间与欧几里得空间的变分自编码器（VAE）在无监督专利权利要求分类与先验检索中的表现，发现超双曲模型能多检索 52% 的跨类别先验技术，并提出可预测方法适用性的领域特征指标。
SAGE 提出了一种在冻结的表示自编码器空间潜变量中直接学习显著因子的方法，并以参考图像的显著表示条件化扩散 Transformer，实现无监督子类型发现与条件化生成，在 Digits-ImageNet、FFHQ 和视网膜 OCT 上显著优于基线。
FB-GDM introduces a fully-Bayesian guided diffusion method for high-dimensional linear inverse problems, eliminating per-task hyperparameter tuning and demonstrating robust performance improvements over existing techniques.
This paper analyzes failures in reconstruction-based unsupervised learning through geometric conditions and proposes new methods like Dynamic Push and Pull to enhance anomaly detection performance.
This paper introduces sheaf regularization to stabilize Decentralized SyncMap for unsupervised continual chunking, achieving higher normalized mutual information and better adaptation to input distribution shifts.
The paper proposes UGCAD, an unsupervised framework using variational graph autoencoders and clustering to detect collective anomalies in IoT network traffic, with experiments on benchmark datasets showing its effectiveness.
This paper introduces an unsupervised, training-free approach to discover prompt-conditional stylistic axes in LLM hidden activations using sampling and PCA, validated through human studies.
DSD is a diffusion-based method for discovering diverse and reusable motor skills in simulated humanoid control, improving upon prior skill discovery techniques with broader behavioral coverage.
This paper presents an unsupervised graph neural network framework for solving the Minimum Dominating Set problem, achieving significant speed improvements and generalization for influence maximization in social networks.
This paper uses 1,744 Landsat and Sentinel-2 satellite images of Robert Smithson's Spiral Jetty to compute multi-feature complexity signatures, revealing that image complexity acts as a leading indicator of lake elevation and climate metrics over 40 years.
This paper shows that hierarchical clustering can simultaneously satisfy scale invariance, richness, and consistency axioms, resolving Kleinberg's Impossibility Theorem for flat clustering. It constructs admissible hierarchical methods and analyzes their diversity and common backbone.
This paper proposes a graph dictionary learning framework that represents graphs as Gaussian distributions using filtered Laplacians and optimal transport distances, achieving competitive performance in graph clustering and classification tasks.
This paper proposes an unsupervised framework combining manifold learning and interpretable graph embeddings to address geometric and interpretability gaps in visual representations, enhancing performance in image retrieval and GCN classification tasks.