Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Scalable Cross-View Sample Alignment for Multi-View Clustering with View Structure Similarity
Authors: Jun Wang, Zhenglai Li, Chang Tang, Suyuan Liu, Hao Yu, Chuan Tang, Miaomiao Li, Xinwang Liu
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The effectiveness of the proposed SSA-MVC algorithm is validated through extensive experiments conducted on eight real-world multi-view datasets. ... Overall, the main contributions of this paper are listed as follows: ... An alternating optimization algorithm is proposed to efficiently solve the model. Its effectiveness is validated through extensive experiments on eight multi-view datasets. ... Section 5 Experiments |
| Researcher Affiliation | Academia | 1National University of Defense Technology, Changsha, China 2Shenzhen Institutes of Advanced Technology, Shenzhen, China 3Huazhong University of Science and Technology, Wuhan, China 4Changsha College, Changsha, China |
| Pseudocode | Yes | A.2 The Pseudo Code of the Proposed Method Algorithm 1 The Algorithm of SSA-MVC. |
| Open Source Code | No | Furthermore, the source code will be released after the review. ... The source code and data will be released after the whole double-blind review. |
| Open Datasets | Yes | To further validate the effectiveness of the proposed method, we conduct experiments on eight real-world multi-view datasets, including Yale, 3sources, MSRCV, 100leaves, HW, Scene, EMNIST, and Hdigit. The detailed summary of them is provided in Appendix A.4. Appendix A.4 Datasets Description: Yale: https://vision.ucsd.edu/content/yale-face-database 3sources: http://mlg.ucd.ie/datasets/3sources.html MSRCV: https://mldta.com/dataset/msrc-v1/ 100leaves: https://archive.ics.uci.edu/ml/datasets/Onehundred+plant+species+leaves+data+set HW: https://archive.ics.uci.edu/ml/datasets/Multiple+Features EMNIST: https://www.nist.gov/itl/products-and-services/emnist-dataset Hdigit: https://cs.nyu.edu/ roweis/data.html |
| Dataset Splits | Yes | To facilitate a fair comparison between the proposed method and existing approaches under the sample non-alignment setting, we fix the sample alignment ratio ρ to 50% in the main experiments. Due to space constraints, results under other alignment ratios are provided in the Appendix and can be found in Tables 7-8 for reference. Appendix A.6 Experimental Results with Varying Sample Alignment Ratios: To further assess the effectiveness of the proposed algorithm under different sample alignment rates, experiments were also conducted at 25% and 75% alignment rates, with detailed results shown in Tables 7-8. |
| Hardware Specification | Yes | All experiments are conducted on a Windows 11 PC equipped with an Intel Core i7-13700F CPU and 64GB RAM. |
| Software Dependencies | No | The paper does not provide specific software names with version numbers for libraries or frameworks used in the experiments. |
| Experiment Setup | Yes | For the proposed method, we conducted a grid search over the set {0.001, 0.01, 0.1, 1, 10, 100, 1000, 10000} to determine the best value for each dataset. The unified latent feature dimension d is set to the number of clusters. Regarding the baseline methods, parameters were tuned according to the ranges provided in their respective publicly available source codes, and the best results were selected in the experiments. To mitigate the influence of randomness on the experimental results, each experiment was repeated 20 times, and the mean and variance of the results are reported. |