arXiv cs.AI· Zhengyun Cheng, Guanwen Zhang, Wei Zhou·· 14 小时前AI 评分18
DiffTR：共享因子条件扩散统一低秩张量恢复
Shared-Factor-Conditioned Diffusion for Low-Rank Tensor Recovery
AI 导读
研究者提出无监督框架 DiffTR，用坐标索引的共享因子条件化紧凑评分网络，通过 score matching 联合学习因子与隐式张量表示，统一处理张量补全、去噪和高光谱超分辨率。该框架将表示学习与任务特定观测建模分离，支持标量和向量输出以覆盖表格张量、图像和视频。实验显示其在多种张量模态和恢复任务上取得有竞争力的重建质量，代码已公开。
正文 · 原文
Abstract:Tensor recovery spans heterogeneous data structures and observation processes, calling for a representation that can adapt across modalities and tasks. We propose DiffTR, an unsupervised framework that unifies tensor completion, denoising, and hyperspectral super-resolution through coordinate-conditioned diffusion. Shared factors indexed by tensor coordinates condition a compact score network, and score matching jointly learns the factors and an implicit tensor representation from the available observations. Scalar and vector-valued outputs accommodate individual entries and entire tubes, enabling a common formulation for tabular tensors, images, and videos. The framework separates representation learning from task-specific observation modeling: reverse diffusion recovers missing values; estimation via Tweedie's formula, soft thresholding, and optional spatial consistency support mixed-noise denoising; and measurement-guided sampling transfers a prior learned from low-resolution hyperspectral observations to high-resolution reconstruction, followed by post-sampling refinement with data consistency and spectral subspace regularization. This separation makes the representation compatible with different corruption models and permits task-specific constraints to be incorporated within the same recovery framework. Experiments across multiple tensor modalities and recovery tasks demonstrate competitive reconstruction quality and the flexibility of a unified conditional generative representation. The code is available at this https URL
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.35368 [cs.CV] |
| (or arXiv:2610.35368v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35368 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhengyun Cheng [view email]
[v1]
Mon, 26 Oct 2026 12:53:53 UTC (34,666 KB)
来源：arXiv cs.AI · arxiv.org