pith. machine review for the scientific record. sign in

arxiv: 2512.21510 · v2 · submitted 2025-12-25 · 💻 cs.LG · cs.CV

Missing Pattern Tree based Decision Grouping and Ensemble for Enhancing Pair Utilization in Deep Incomplete Multi-View Clustering

Jie Xu , Wenyuan Yang , Yazhou Ren , Lifang He , Philip S. Yu , Xiaofeng Zhu This is my paper

Pith reviewed 2026-05-16 20:00 UTC · model grok-4.3

classification 💻 cs.LG cs.CV
keywords incomplete multi-view clusteringmissing pattern treedecision groupingensemble learningknowledge distillationpair utilizationdeep clustering
0
0 comments X

The pith

A missing-pattern tree groups data by missing patterns to make fuller use of available multi-view pairs in clustering.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Inconsistent missing patterns across views cause many usable data pairs to be under-exploited during clustering. The method first builds a tree that partitions instances into groups sharing identical missing patterns, then runs multi-view clustering inside each group. An uncertainty-weighted ensemble combines the group results while suppressing weak decisions, and a distillation step passes the ensemble output back to improve the separate view models through added consistency and discrimination objectives. A reader would care because the design targets the specific waste of available pairs without imputation, potentially raising accuracy when real-world missingness is irregular.

Core claim

The paper establishes that a missing-pattern tree can partition the data into multiple decision sets according to missing patterns, allowing multi-view clustering to be performed within each set, after which an uncertainty-weighted ensemble aggregates the decisions and an ensemble-to-individual distillation transfers knowledge to enhance view-specific models via cross-view consistency and inter-cluster discrimination losses.

What carries the argument

Missing-pattern tree that partitions data into decision sets by their missing patterns, enabling per-set clustering followed by uncertainty-weighted ensemble aggregation and knowledge distillation.

If this is right

  • Available but incomplete pairs across views get used inside their matching missing-pattern groups instead of being ignored.
  • Uncertainty weights downplay unreliable decisions from groups with few complete pairs.
  • Distillation from the ensemble back to individual models strengthens cross-view consistency and cluster separation.
  • Theoretical analysis backs the grouping and ensemble steps as contributors to higher overall performance.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Pattern-based grouping could apply to other incomplete-data settings such as sensor fusion where missingness follows observable rules.
  • Avoiding imputation steps may simplify pipelines for real multi-view applications with irregular missing entries.
  • The tree structure might be tested for adaptability when new missing patterns emerge over time in streaming data.

Load-bearing premise

That grouping instances by shared missing patterns produces stable within-group clusters whose ensemble combination improves overall results without adding new biases from small or uneven groups.

What would settle it

Experiments on the same benchmarks that show no gain in standard clustering metrics such as accuracy or normalized mutual information when the missing-pattern grouping and ensemble are removed.

Figures

Figures reproduced from arXiv: 2512.21510 by Jie Xu, Lifang He, Philip S. Yu, Wenyuan Yang, Xiaofeng Zhu, Yazhou Ren.

Figure 1
Figure 1. Figure 1: Framework Overview of Our TreeEIC. First, view-specific models generate sample embeddings via autoencoders. (a) The embeddings are grouped into multiple decision subsets according to missing patterns. Samples in each subset share the consistent missing pattern, allowing to obtain their clustering decisions in one feature space. (b) Clustering decisions from all subsets are aligned and weighted based on unc… view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the missing patterns in incomplete multi-view data. Numerous incomplete multi-view data exhibiting inconsistent missing patterns can be grouped into multiple decision sets. The missing pattern of samples in each decision set exhibits consistency and thus the pair relationship in these samples can be exploited to improve IMVC. of the v-th view binary mask a v ∈ A. To systematically represent… view at source ↗
Figure 3
Figure 3. Figure 3: ACC vs. Missing Rate on AWA-7 and ModelNet40. When the missing rate τ = 1.0, i.e., the highly inconsistent miss￾ing patterns, we can observe most IMVC methods have heavy performance degradation while our method TreeEIC is still robust. 6 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: ACC vs. Loss on HandWritten and Caltech101-7. loss. Compared with the basic result, we can observe that both Lcons and Ldisc can individually improve the model performance. For example, on the HandWritten and AWA-7 datasets, the introduction of Lcons brings accuracy improve￾ments of 55.67% and 31.85%, respectively, while Ldisc con￾tributes additional gains of 57.12% and 30.15%. This is because the cross-vi… view at source ↗
Figure 5
Figure 5. Figure 5: Representation learned without (first row) or with (second row) our ensemble-to-individual distillation on 6 views of HandWritten. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Parameter analysis on Caltech101-7 and HandWritten. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 1
Figure 1. Figure 1: ACC vs. Missing Rate across six datasets. When the missing rate τ = 1.0, i.e., the highly inconsistent missing patterns, we can observe that most IMVC methods have heavy performance degradation while our method TreeEIC is still robust [PITH_FULL_IMAGE:figures/full_fig_p015_1.png] view at source ↗
read the original abstract

Real-world multi-view data often exhibit highly inconsistent missing patterns, posing significant challenges for incomplete multi-view clustering (IMVC). Although existing IMVC methods have made progress from both imputation-based and imputation-free routes, they largely overlook the issue of pair underutilization. Specifically, inconsistent missing patterns prevent incomplete but available multi-view pairs from being fully exploited, thereby limiting the model performance. To address this limitation, we propose a novel missing-pattern tree based IMVC framework. Specifically, to fully leverage available multi-view pairs, we first introduce a missing-pattern tree model to group data into multiple decision sets according to their missing patterns, and then perform multi-view clustering within each set. Furthermore, a multi-view decision ensemble module is proposed to aggregate clustering results across all decision sets. This module infers uncertainty-based weights to suppress unreliable clustering decisions and produce robust outputs. Finally, we develop an ensemble-to-individual knowledge distillation module module, which transfers ensemble knowledge to view-specific clustering models. This design enables mutual enhancement between ensemble and individual modules by optimizing cross-view consistency and inter-cluster discrimination losses. Extensive theoretical analysis supports our key designs, and empirical experiments on multiple benchmark datasets demonstrate that our method effectively mitigates the pair underutilization issue and achieve superior IMVC performance.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper proposes a missing-pattern tree based IMVC framework to address pair underutilization arising from inconsistent missing patterns in multi-view data. It groups samples into decision sets by missing-pattern identity, performs per-set multi-view clustering, aggregates results via an uncertainty-weighted ensemble module, and applies ensemble-to-individual knowledge distillation to mutually enhance the modules through cross-view consistency and inter-cluster discrimination losses. The approach is supported by theoretical analysis and empirical results on benchmark datasets.

Significance. If the grouping, weighting, and distillation components prove robust, the work could meaningfully advance IMVC by better exploiting available incomplete pairs rather than discarding them, offering a practical alternative to imputation-heavy or fully imputation-free baselines in settings with highly variable missingness.

major comments (2)
  1. [Abstract] Abstract and method description: the core grouping step partitions data by missing-pattern identity before per-set clustering, yet the manuscript provides no explicit mechanism (e.g., minimum-set-size threshold, merging rule, or regularization) to prevent fragmentation into tiny decision sets when patterns are highly inconsistent; this directly threatens the reliability of the subsequent cross-view consistency and discrimination losses on small-n subsets.
  2. [Abstract] Abstract: the claim that uncertainty-based weights 'suppress unreliable clustering decisions' is load-bearing for the ensemble module, but no derivation or empirical verification is shown that these weights remain stable or informative when base clusterings are themselves estimated from very small decision sets.
minor comments (1)
  1. The abstract refers to 'extensive theoretical analysis' supporting the key designs; the manuscript should include a dedicated section or appendix that states the precise assumptions and lemmas rather than leaving them implicit.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback highlighting potential vulnerabilities in decision-set fragmentation and uncertainty-weight stability. We address each major comment below and will revise the manuscript accordingly to strengthen the presentation and robustness of the missing-pattern tree framework.

read point-by-point responses

  1. Referee: [Abstract] Abstract and method description: the core grouping step partitions data by missing-pattern identity before per-set clustering, yet the manuscript provides no explicit mechanism (e.g., minimum-set-size threshold, merging rule, or regularization) to prevent fragmentation into tiny decision sets when patterns are highly inconsistent; this directly threatens the reliability of the subsequent cross-view consistency and discrimination losses on small-n subsets.

    Authors: We agree that an explicit safeguard is desirable to ensure decision sets remain large enough for stable estimation of cross-view consistency and discrimination losses. The missing-pattern tree is designed to group samples by exact missing-pattern identity precisely to maximize utilization of available pairs within each group. On the benchmark datasets, the resulting sets proved sufficiently large. To address the general case of highly inconsistent patterns, we will introduce a minimum-set-size threshold in the revised method section: sets below the threshold will be merged with the nearest compatible pattern group (using Hamming distance on missing masks) or regularized via a shared prototype. We will also add a sensitivity study on the threshold value and report its effect on clustering metrics. revision: yes

  2. Referee: [Abstract] Abstract: the claim that uncertainty-based weights 'suppress unreliable clustering decisions' is load-bearing for the ensemble module, but no derivation or empirical verification is shown that these weights remain stable or informative when base clusterings are themselves estimated from very small decision sets.

    Authors: The uncertainty weights are derived from the variance of soft cluster assignments produced by the per-set clustering module; higher variance yields lower weight, thereby suppressing unreliable decisions. The theoretical analysis already shows that this weighting improves ensemble robustness under the assumption of independent set-wise errors. We acknowledge that explicit verification for the small-set regime was not provided. In the revision we will (i) add a short derivation bounding the weight variance as a function of set size and (ii) include new empirical plots that display weight distributions and downstream NMI/ACC as decision-set size varies, confirming that the weights remain informative down to moderate sizes once the minimum-set threshold is applied. revision: yes

Circularity Check

0 steps flagged

No circularity: new framework construction with independent empirical support

full rationale

The paper proposes a missing-pattern tree for grouping samples by missing patterns, followed by per-group multi-view clustering, uncertainty-weighted ensemble, and ensemble-to-individual distillation. No equations, fitted parameters, or self-cited uniqueness theorems are shown that reduce any claimed prediction or performance gain to a re-expression of the inputs by construction. The derivation chain consists of explicit algorithmic steps (tree construction, per-set clustering, weighting, distillation) whose validity is asserted via theoretical analysis and benchmark experiments rather than definitional equivalence or load-bearing self-citation. This is the normal case of a self-contained algorithmic contribution.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no explicit free parameters, axioms, or invented entities; the framework is described conceptually without numerical fitting details or unstated background assumptions.

pith-pipeline@v0.9.0 · 5542 in / 1139 out tokens · 31464 ms · 2026-05-16T20:00:07.957863+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

74 extracted references · 74 canonical work pages · 4 internal anchors

  1. [1]

    Adaptive cluster ensemble selection

    Javad Azimi and Xiaoli Z Fern. Adaptive cluster ensemble selection. InIjcai, pages 992–997, 2009. 12

    work page 2009

  2. [2]

    Repre- sentation learning: A review and new perspectives.TPAMI, 35(8):1798–1828, 2013

    Yoshua Bengio, Aaron Courville, and Pascal Vincent. Repre- sentation learning: A review and new perspectives.TPAMI, 35(8):1798–1828, 2013. 12

    work page 2013

  3. [3]

    Incomplete multi-view clustering with multiple imputation and ensemble clustering.Applied Intelli- gence, 52(13):14811–14821, 2022

    Guoqing Chao, Songtao Wang, Shiming Yang, Chunshan Li, and Dianhui Chu. Incomplete multi-view clustering with multiple imputation and ensemble clustering.Applied Intelli- gence, 52(13):14811–14821, 2022. 12

    work page 2022

  4. [4]

    Incomplete con- trastive multi-view clustering with high-confidence guiding

    Guoqing Chao, Yi Jiang, and Dianhui Chu. Incomplete con- trastive multi-view clustering with high-confidence guiding. InProceedings of the AAAI conference on artificial intelli- gence, pages 11221–11229, 2024. 1, 2, 5, 6, 14

    work page 2024

  5. [5]

    Global graph propagation with hierarchical information trans- fer for incomplete contrastive multi-view clustering

    Guoqing Chao, Kaixin Xu, Xijiong Xie, and Yongyong Chen. Global graph propagation with hierarchical information trans- fer for incomplete contrastive multi-view clustering. InPro- ceedings of the AAAI Conference on Artificial Intelligence, pages 15713–15721, 2025. 5, 6, 14

    work page 2025

  6. [6]

    Multi-view clustering via canonical cor- relation analysis

    Kamalika Chaudhuri, Sham M Kakade, Karen Livescu, and Karthik Sridharan. Multi-view clustering via canonical cor- relation analysis. InProceedings of the 26th annual interna- tional conference on machine learning, pages 129–136, 2009. 1

    work page 2009

  7. [7]

    Deep multiview clustering by contrasting cluster assignments

    Jie Chen, Hua Mao, Wai Lok Woo, and Xi Peng. Deep multiview clustering by contrasting cluster assignments. In Proceedings of the IEEE/CVF international conference on computer vision, pages 16752–16761, 2023. 12

    work page 2023

  8. [8]

    Imputation-free and alignment-free: Incomplete multi-view clustering driven by consensus semantic learning

    Yuzhuo Dai, Jiaqi Jin, Zhibin Dong, Siwei Wang, Xinwang Liu, En Zhu, Xihong Yang, Xinbiao Gan, and Yu Feng. Imputation-free and alignment-free: Incomplete multi-view clustering driven by consensus semantic learning. InPro- ceedings of the Computer Vision and Pattern Recognition Conference, pages 5071–5081, 2025. 2, 5, 6, 12, 13, 14

    work page 2025

  9. [9]

    Ensemble learning.The handbook of brain theory and neural networks, 2(1):110–125, 2002

    Thomas G Dietterich et al. Ensemble learning.The handbook of brain theory and neural networks, 2(1):110–125, 2002. 2

    work page 2002

  10. [10]

    Selective cross-view topology for deep incomplete multi-view clustering.IEEE Transactions on Image Processing, 2025

    Zhibin Dong, Dayu Hu, Jiaqi Jin, Siwei Wang, Xinwang Liu, and En Zhu. Selective cross-view topology for deep incomplete multi-view clustering.IEEE Transactions on Image Processing, 2025. 1

    work page 2025

  11. [11]

    Learning gener- ative visual models from few training examples: An incre- mental bayesian approach tested on 101 object categories

    Li Fei-Fei, Rob Fergus, and Pietro Perona. Learning gener- ative visual models from few training examples: An incre- mental bayesian approach tested on 101 object categories. In CVPR, pages 178–178, 2004. 5, 12, 14

    work page 2004

  12. [12]

    Cluster ensemble selection.Sta- tistical Analysis and Data Mining: The ASA Data Science Journal, 1(3):128–141, 2008

    Xiaoli Z Fern and Wei Lin. Cluster ensemble selection.Sta- tistical Analysis and Data Mining: The ASA Data Science Journal, 1(3):128–141, 2008. 12

    work page 2008

  13. [13]

    Comment-based multi-view clustering of web 2.0 items

    Xiangnan He, Min-Yen Kan, Peichu Xie, and Xiao Chen. Comment-based multi-view clustering of web 2.0 items. In Proceedings of the 23rd international conference on World wide web, pages 771–782, 2014. 1

    work page 2014

  14. [14]

    Reducing the dimensionality of data with neural networks.Science, 313 (5786):504–507, 2006

    Geoffrey E Hinton and Ruslan R Salakhutdinov. Reducing the dimensionality of data with neural networks.Science, 313 (5786):504–507, 2006. 12

    work page 2006

  15. [15]

    One-pass incomplete multi- view clustering

    Menglei Hu and Songcan Chen. One-pass incomplete multi- view clustering. InProceedings of the AAAI conference on artificial intelligence, pages 3838–3845, 2019. 1

    work page 2019

  16. [16]

    Multi-view spectral clustering via integrating nonnegative embedding and spectral embedding.Information Fusion, 55: 251–259, 2020

    Zhanxuan Hu, Feiping Nie, Rong Wang, and Xuelong Li. Multi-view spectral clustering via integrating nonnegative embedding and spectral embedding.Information Fusion, 55: 251–259, 2020. 5, 14

    work page 2020

  17. [17]

    Fast multi-view clustering via ensembles: Towards scalability, superiority, and simplicity.IEEE Transactions on Knowledge and Data Engineering, 35(11):11388–11402, 2023

    Dong Huang, Chang-Dong Wang, and Jian-Huang Lai. Fast multi-view clustering via ensembles: Towards scalability, superiority, and simplicity.IEEE Transactions on Knowledge and Data Engineering, 35(11):11388–11402, 2023. 12

    work page 2023

  18. [18]

    Multispectralnet: Spectral clustering using deep neural network for multi-view data.IEEE Transactions on Compu- tational Social Systems, 6(4):749–760, 2019

    Shuning Huang, Kaoru Ota, Mianxiong Dong, and Fanzhang Li. Multispectralnet: Spectral clustering using deep neural network for multi-view data.IEEE Transactions on Compu- tational Social Systems, 6(4):749–760, 2019. 1

    work page 2019

  19. [19]

    Bagging-based spectral clustering ensemble selection.Pattern Recognition Letters, 32(10):1456–1467, 2011

    Jianhua Jia, Xuan Xiao, Bingxiang Liu, and Licheng Jiao. Bagging-based spectral clustering ensemble selection.Pattern Recognition Letters, 32(10):1456–1467, 2011. 12

    work page 2011

  20. [20]

    Improving the hungarian assignment algorithm.Operations Research Letters, 5(4): 171–175, 1986

    Roy Jonker and Ton V olgenant. Improving the hungarian assignment algorithm.Operations Research Letters, 5(4): 171–175, 1986. 4

    work page 1986

  21. [21]

    Pointnet with kan versus pointnet with mlp for 3d classification and segmentation of point sets.arXiv preprint arXiv:2410.10084, 2024

    Ali Kashefi. Pointnet with kan versus pointnet with mlp for 3d classification and segmentation of point sets.arXiv preprint arXiv:2410.10084, 2024. 13

    work page arXiv 2024

  22. [22]

    Rethinking multi-view representation learning via distilled disentangling

    Guanzhou Ke, Bo Wang, Xiaoli Wang, and Shengfeng He. Rethinking multi-view representation learning via distilled disentangling. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 26774– 26783, 2024. 1

    work page 2024

  23. [23]

    Adam: A Method for Stochastic Optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980,

    work page internal anchor Pith review Pith/arXiv arXiv

  24. [24]

    A co-training approach for multi-view spectral clustering

    Abhishek Kumar and Hal Daum ´e. A co-training approach for multi-view spectral clustering. InProceedings of the 28th international conference on machine learning (ICML-11), pages 393–400, 2011. 1

    work page 2011

  25. [25]

    Co-regularized multi-view spectral clustering.Advances in neural informa- tion processing systems, 24, 2011

    Abhishek Kumar, Piyush Rai, and Hal Daume. Co-regularized multi-view spectral clustering.Advances in neural informa- tion processing systems, 24, 2011. 1

    work page 2011

  26. [26]

    Backpropagation applied to handwritten zip code recognition.Neural computation, 1(4):541–551, 1989

    Yann LeCun, Bernhard Boser, John S Denker, Donnie Hen- derson, Richard E Howard, Wayne Hubbard, and Lawrence D Jackel. Backpropagation applied to handwritten zip code recognition.Neural computation, 1(4):541–551, 1989. 5, 12, 14

    work page 1989

  27. [27]

    Incomplete multi-view cluster- ing via prototype-based imputation.arXiv preprint arXiv:2301.11045, 2023

    Haobin Li, Yunfan Li, Mouxing Yang, Peng Hu, Dezhong Peng, and Xi Peng. Incomplete multi-view cluster- ing via prototype-based imputation.arXiv preprint arXiv:2301.11045, 2023. 1, 5, 6, 14

    work page arXiv 2023

  28. [28]

    Incomplete multi- view clustering with joint partition and graph learning.IEEE Transactions on Knowledge and Data Engineering, 35(1): 589–602, 2021

    Lusi Li, Zhiqiang Wan, and Haibo He. Incomplete multi- view clustering with joint partition and graph learning.IEEE Transactions on Knowledge and Data Engineering, 35(1): 589–602, 2021. 1

    work page 2021

  29. [29]

    Partial multi- view clustering

    Shao-Yuan Li, Yuan Jiang, and Zhi-Hua Zhou. Partial multi- view clustering. InProceedings of the AAAI conference on artificial intelligence, 2014. 1

    work page 2014

  30. [30]

    A survey of multi-view representation learning.IEEE transactions on knowledge and data engineering, 31(10):1863–1883, 2018

    Yingming Li, Ming Yang, and Zhongfei Zhang. A survey of multi-view representation learning.IEEE transactions on knowledge and data engineering, 31(10):1863–1883, 2018. 1

    work page 2018

  31. [31]

    Deep adversarial multi-view clustering network

    Zhaoyang Li, Qianqian Wang, Zhiqiang Tao, Quanxue Gao, Zhaohua Yang, et al. Deep adversarial multi-view clustering network. InIJCAI, page 4, 2019. 12 9

    work page 2019

  32. [32]

    Jointly Deep Multi-View Learning for Clustering Analysis

    Bingqian Lin, Yuan Xie, Yanyun Qu, Cuihua Li, and Xiao- dan Liang. Jointly deep multi-view learning for clustering analysis.arXiv preprint arXiv:1808.06220, 2018. 12

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  33. [33]

    Completer: Incomplete multi-view clustering via contrastive prediction

    Yijie Lin, Yuanbiao Gou, Zitao Liu, Boyun Li, Jiancheng Lv, and Xi Peng. Completer: Incomplete multi-view clustering via contrastive prediction. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11174–11183, 2021. 12

    work page 2021

  34. [34]

    Localized sparse incomplete multi-view clustering

    Chengliang Liu, Zhihao Wu, Jie Wen, Yong Xu, and Chao Huang. Localized sparse incomplete multi-view clustering. IEEE Transactions on Multimedia, 25:5539–5551, 2022. 1

    work page 2022

  35. [35]

    Late fusion incomplete multi-view clustering.TPAMI, 41(10):2410–2423, 2018

    Xinwang Liu, Xinzhong Zhu, Miaomiao Li, Lei Wang, Chang Tang, Jianping Yin, Dinggang Shen, Huaimin Wang, and Wen Gao. Late fusion incomplete multi-view clustering.TPAMI, 41(10):2410–2423, 2018. 12

    work page 2018

  36. [36]

    Multivariate observations

    J MacQueen. Multivariate observations. InProceedings ofthe 5th Berkeley Symposium on Mathematical Statisticsand Probability, pages 281–297, 1967. 4

    work page 1967

  37. [37]

    Representation Learning with Contrastive Predictive Coding

    Aaron van den Oord, Yazhe Li, and Oriol Vinyals. Repre- sentation learning with contrastive predictive coding.arXiv preprint arXiv:1807.03748, 2018. 4

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  38. [38]

    Variational relational point completion network

    Liang Pan, Xinyi Chen, Zhongang Cai, Junzhe Zhang, Haiyu Zhao, Shuai Yi, and Ziwei Liu. Variational relational point completion network. InProceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 8524–8533, 2021. 5, 13, 14

    work page 2021

  39. [39]

    Pointnet: Deep learning on point sets for 3d classification and segmentation

    Charles R Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas. Pointnet: Deep learning on point sets for 3d classification and segmentation. InProceedings of the IEEE conference on computer vision and pattern recognition, pages 652–660,

    work page

  40. [40]

    An embarrass- ingly simple approach to zero-shot learning

    Bernardino Romera-Paredes and Philip Torr. An embarrass- ingly simple approach to zero-shot learning. InInternational conference on machine learning, pages 2152–2161. PMLR,

    work page

  41. [41]

    Cluster ensembles—a knowledge reuse framework for combining multiple partitions

    Alexander Strehl and Joydeep Ghosh. Cluster ensembles—a knowledge reuse framework for combining multiple partitions. Journal of machine learning research, 3(Dec):583–617, 2002. 12

    work page 2002

  42. [42]

    Incom- plete multi-view learning: Review, analysis, and prospects

    Jingjing Tang, Qingqing Yi, Saiji Fu, and Yingjie Tian. Incom- plete multi-view learning: Review, analysis, and prospects. Applied Soft Computing, 153:111278, 2024. 1

    work page 2024

  43. [43]

    From ensemble clustering to multi-view clustering

    Zhiqiang Tao, Hongfu Liu, Sheng Li, Zhengming Ding, and Yun Fu. From ensemble clustering to multi-view clustering. InIJCAI, 2017. 12

    work page 2017

  44. [44]

    Multigraph spectral clustering for joint content delivery and scheduling in beam-free satellite communications

    Miguel Angel V´azquez and Ana I P´erez-Neira. Multigraph spectral clustering for joint content delivery and scheduling in beam-free satellite communications. InICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 8802–8806. IEEE, 2020. 1

    work page 2020

  45. [45]

    Fast continual multi-view clustering with incomplete views.IEEE Transactions on Image Processing, 33:2995–3008, 2024

    Xinhang Wan, Bin Xiao, Xinwang Liu, Jiyuan Liu, Weixuan Liang, and En Zhu. Fast continual multi-view clustering with incomplete views.IEEE Transactions on Image Processing, 33:2995–3008, 2024. 1

    work page 2024

  46. [46]

    Tensor completion framework by graph refinement for incomplete multi-view clustering.IEEE Transactions on Multimedia, 2025

    Huibing Wang, Yawei Chen, Mingze Yao, Wenzhe Liu, Jinjia Peng, and Xianping Fu. Tensor completion framework by graph refinement for incomplete multi-view clustering.IEEE Transactions on Multimedia, 2025. 1

    work page 2025

  47. [47]

    Generative partial multi-view clustering with adaptive fusion and cycle consistency.IEEE Transac- tions on Image Processing, 30:1771–1783, 2021

    Qianqian Wang, Zhengming Ding, Zhiqiang Tao, Quanxue Gao, and Yun Fu. Generative partial multi-view clustering with adaptive fusion and cycle consistency.IEEE Transac- tions on Image Processing, 30:1771–1783, 2021. 1

    work page 2021

  48. [48]

    Highly-efficient incomplete large-scale multi-view clustering with consensus bipartite graph

    Siwei Wang, Xinwang Liu, Li Liu, Wenxuan Tu, Xinzhong Zhu, Jiyuan Liu, Sihang Zhou, and En Zhu. Highly-efficient incomplete large-scale multi-view clustering with consensus bipartite graph. InCVPR, pages 9776–9785, 2022. 1

    work page 2022

  49. [49]

    Deep incomplete multi-view multiple clusterings

    Shaowei Wei, Jun Wang, Guoxian Yu, Carlotta Domeniconi, and Xiangliang Zhang. Deep incomplete multi-view multiple clusterings. In2020 IEEE International Conference on Data Mining (ICDM), pages 651–660. IEEE, 2020. 1

    work page 2020

  50. [50]

    Incomplete multiview spectral clustering with adaptive graph learning.IEEE trans- actions on cybernetics, 50(4):1418–1429, 2018

    Jie Wen, Yong Xu, and Hong Liu. Incomplete multiview spectral clustering with adaptive graph learning.IEEE trans- actions on cybernetics, 50(4):1418–1429, 2018. 1

    work page 2018

  51. [51]

    In- complete multi-view clustering via graph regularized matrix factorization

    Jie Wen, Zheng Zhang, Yong Xu, and Zuofeng Zhong. In- complete multi-view clustering via graph regularized matrix factorization. InProceedings of the European conference on computer vision (ECCV) workshops, pages 0–0, 2018. 1

    work page 2018

  52. [52]

    Adaptive graph completion based incomplete multi-view clustering.IEEE Transactions on Multimedia, 23:2493–2504, 2020

    Jie Wen, Ke Yan, Zheng Zhang, Yong Xu, Junqian Wang, Lunke Fei, and Bob Zhang. Adaptive graph completion based incomplete multi-view clustering.IEEE Transactions on Multimedia, 23:2493–2504, 2020. 1

    work page 2020

  53. [53]

    DIMC-net: Deep incomplete multi-view clustering network

    Jie Wen, Zheng Zhang, Zhao Zhang, Zhihao Wu, Lunke Fei, Yong Xu, and Bob Zhang. DIMC-net: Deep incomplete multi-view clustering network. InACM MM, pages 3753– 3761, 2020. 1

    work page 2020

  54. [54]

    Imputation- free incomplete multi-view clustering via knowledge distil- lation

    Benyu Wu, Wei Du, Jun Wang, and Guoxian Yu. Imputation- free incomplete multi-view clustering via knowledge distil- lation. InProceedings of the Computer Vision and Pattern Recognition Conference, pages 5071–5081, 2025. 1, 5, 13

    work page 2025

  55. [55]

    Relationship completion for incomplete multi- view clustering.Neural Networks, page 107791, 2025

    Minghong Wu, Jihua Zhu, Wenbiao Yan, Bin Chen, and Qing- hai Zheng. Relationship completion for incomplete multi- view clustering.Neural Networks, page 107791, 2025. 1

    work page 2025

  56. [56]

    3d shapenets: A deep representation for volumetric shapes

    Zhirong Wu, Shuran Song, Aditya Khosla, Fisher Yu, Lin- guang Zhang, Xiaoou Tang, and Jianxiong Xiao. 3d shapenets: A deep representation for volumetric shapes. InProceedings of the IEEE conference on computer vision and pattern recog- nition, pages 1912–1920, 2015. 5, 12, 14

    work page 1912

  57. [57]

    Multi-view clustering ensembles

    Xin Xie and Shiliang Sun. Multi-view clustering ensembles. InProceedings of the 2013 International Conference on Ma- chine Learning and Cybernetics, pages 51–56. IEEE, 2013. 12

    work page 2013

  58. [58]

    A Survey on Multi-view Learning

    Chang Xu, Dacheng Tao, and Chao Xu. A survey on multi- view learning.arXiv preprint arXiv:1304.5634, 2013. 1

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  59. [59]

    Deep incomplete multi- view clustering via mining cluster complementarity

    Jie Xu, Chao Li, Yazhou Ren, Liang Peng, Yujie Mo, Xi- aoshuang Shi, and Xiaofeng Zhu. Deep incomplete multi- view clustering via mining cluster complementarity. InPro- ceedings of the AAAI conference on artificial intelligence, pages 8761–8769, 2022. 1, 12

    work page 2022

  60. [60]

    Self- supervised discriminative feature learning for deep multi- view clustering.IEEE Transactions on Knowledge and Data Engineering, 35(7):7470–7482, 2022

    Jie Xu, Yazhou Ren, Huayi Tang, Zhimeng Yang, Lili Pan, Yang Yang, Xiaorong Pu, Philip S Yu, and Lifang He. Self- supervised discriminative feature learning for deep multi- view clustering.IEEE Transactions on Knowledge and Data Engineering, 35(7):7470–7482, 2022. 12 10

    work page 2022

  61. [61]

    Adaptive feature projec- tion with distribution alignment for deep incomplete multi- view clustering.IEEE Transactions on Image Processing, 32: 1354–1366, 2023

    Jie Xu, Chao Li, Liang Peng, Yazhou Ren, Xiaoshuang Shi, Heng Tao Shen, and Xiaofeng Zhu. Adaptive feature projec- tion with distribution alignment for deep incomplete multi- view clustering.IEEE Transactions on Image Processing, 32: 1354–1366, 2023. 1, 5, 6, 14

    work page 2023

  62. [62]

    Investigating and mitigating the side effects of noisy views for self-supervised clustering algorithms in practical multi-view scenarios

    Jie Xu, Yazhou Ren, Xiaolong Wang, Lei Feng, Zheng Zhang, Gang Niu, and Xiaofeng Zhu. Investigating and mitigating the side effects of noisy views for self-supervised clustering algorithms in practical multi-view scenarios. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 22957–22966, 2024. 3, 4, 5, 13

    work page 2024

  63. [63]

    Clustering-induced adaptive structure enhancing network for incomplete multi-view data

    Zhe Xue, Junping Du, Changwei Zheng, Jie Song, Wenqi Ren, and Meiyu Liang. Clustering-induced adaptive structure enhancing network for incomplete multi-view data. InIJCAI, pages 3235–3241, 2021. 1

    work page 2021

  64. [64]

    Deep incomplete multi-view clustering via dynamic impu- tation and triple alignment with dual optimization.IEEE Transactions on Circuits and Systems for Video Technology,

    Weiqing Yan, Kanglong Liu, Wujie Zhou, and Chang Tang. Deep incomplete multi-view clustering via dynamic impu- tation and triple alignment with dual optimization.IEEE Transactions on Circuits and Systems for Video Technology,

    work page

  65. [65]

    Deep multiview collaborative clustering.IEEE Transactions on Neural Networks and Learning Systems, 34(1):516–526,

    Xu Yang, Cheng Deng, Zhiyuan Dang, and Dacheng Tao. Deep multiview collaborative clustering.IEEE Transactions on Neural Networks and Learning Systems, 34(1):516–526,

    work page

  66. [66]

    Multi-view clustering: A survey

    Yan Yang and Hao Wang. Multi-view clustering: A survey. Big Data Mining and Analytics, 1(2):83–107, 2018. 1

    work page 2018

  67. [67]

    Incomplete multi- view clustering via subspace learning

    Qiyue Yin, Shu Wu, and Liang Wang. Incomplete multi- view clustering via subspace learning. InProceedings of the 24th ACM international on conference on information and knowledge management, pages 383–392, 2015. 1

    work page 2015

  68. [68]

    Prototype matching learning for incomplete multi-view clustering.IEEE Transactions on Image Processing, 2025

    Honglin Yuan, Yuan Sun, Fei Zhou, Jing Wen, Shihua Yuan, Xiaojian You, and Zhenwen Ren. Prototype matching learning for incomplete multi-view clustering.IEEE Transactions on Image Processing, 2025. 1, 5, 13

    work page 2025

  69. [69]

    Semantic invariant multi-view cluster- ing with fully incomplete information.IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(4):2139–2150,

    Pengxin Zeng, Mouxing Yang, Yiding Lu, Changqing Zhang, Peng Hu, and Xi Peng. Semantic invariant multi-view cluster- ing with fully incomplete information.IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(4):2139–2150,

    work page

  70. [70]

    Enhanced tensor low-rank and sparse representation recovery for incomplete multi-view clustering

    Chao Zhang, Huaxiong Li, Wei Lv, Zizheng Huang, Yang Gao, and Chunlin Chen. Enhanced tensor low-rank and sparse representation recovery for incomplete multi-view clustering. InProceedings of the AAAI conference on artificial intelli- gence, pages 11174–11182, 2023. 1

    work page 2023

  71. [71]

    Preventing dimensional collapse of incomplete multi-view clustering via direct contrastive learning.arXiv preprint arXiv:2303.12241, 2023

    Kaiwu Zhang, Shiqiang Du, Baokai Liu, and Shengxia Gao. Preventing dimensional collapse of incomplete multi-view clustering via direct contrastive learning.arXiv preprint arXiv:2303.12241, 2023. 1

    work page arXiv 2023

  72. [72]

    Incomplete multi-view clustering via diffusion contrastive generation

    Yuanyang Zhang, Yijie Lin, Weiqing Yan, Li Yao, Xinhang Wan, Guangyuan Li, Chao Zhang, Guanzhou Ke, and Jie Xu. Incomplete multi-view clustering via diffusion contrastive generation. InProceedings of the AAAI Conference on Artifi- cial Intelligence, pages 22650–22658, 2025. 5, 6, 14

    work page 2025

  73. [73]

    Dual alignment self-supervised incomplete multi-view subspace clustering network.IEEE Signal Processing Letters, 28:2122– 2126, 2021

    Liang Zhao, Jie Zhang, Qiuhao Wang, and Zhikui Chen. Dual alignment self-supervised incomplete multi-view subspace clustering network.IEEE Signal Processing Letters, 28:2122– 2126, 2021. 1

    work page 2021

  74. [74]

    N/A” indicates that the method cannot run in the case ofρ= 1.0. “OOM

    Zhi-Hua Zhou and Wei Tang. Clusterer ensemble.Knowledge- Based Systems, 19(1):77–83, 2006. 12 11 A. Related Work Deep Multi-View Clustering.Unlike traditional MVC methods that rely on linear transformations, deep Multi- View Clustering[32] (deep MVC) leverages deep neural net- works to learn nonlinear feature representations and cluster- ing assignments i...

    work page 2006