Pith. sign in

REVIEW 3 major objections 5 minor 33 references

Trusted Mamba Contrastive Network for Multi-View Clustering

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Multi-view clustering becomes trustworthy when fusion is selective and contrastive learning follows clusters rather than instances; the paper reports 98.53% ACC on Cifar100, about 32 points above the best baseline.

desk verdict A promising Mamba-based multi-view clustering architecture, but the central SOTA claim currently rests on a likely invalid loss equation and single-run numbers that need verification. read the letter →

arxiv 2412.16487 v2 pith:KOBDU2OL submitted 2024-12-21 cs.CV

Jian Zhu , Xin Zou , Lei Liu , Zhangmin Huang , Ying Zhang , Chang Tang , Li-Rong Dai This is my paper · ORCID
classification cs.CV
keywords multi-viewclusteringfusioncontrastivelearningselectivestatespacemodelMambatrusteddeepunsupervisedrepresentation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Deep multi-view clustering methods typically fuse views by averaging or concatenating embeddings, and align views by instance-level contrastive learning, so noise in one view or a sample whose views disagree can pull the consensus representation away from the true clusters. The paper tries to fix both failure modes at once: it introduces a Trusted Mamba Fusion Network (TMFN) whose selective state-space gate filters noisy and redundant view information, and an Average-similarity Contrastive Learning (AsCL) loss that replaces instance-level positive pairs with cluster-level similarity weighting. On four benchmarks the method reports state-of-the-art accuracy, including 98.53% ACC on Cifar100, about 32 percentage points above the best compared method, and ablations attribute the gains to both modules. The central claim is that selective fusion plus cluster-aware contrastive alignment is enough to make multi-view fusion trustworthy.

What carries the argument

The load-bearing machinery is the pair (TMFN, AsCL). TMFN (Trusted Mamba Fusion Network) turns each view embedding into a sequence of tokens, concatenates the view sequences into one global sequence, and feeds it to a selective state-space model in which the matrices $B$ and $C$ depend on the input (Eq. 8), so the hidden state acts as a learned gate that filters which view information passes through; a second branch applies a gated nonlinearity, and the two are multiplied elementwise (Eq. 9) before conversion back to a fused vector. AsCL (Average-similarity Contrastive Learning) computes $S_{ij}$ as the average cosine similarity between the $i$-th and $j$-th samples over all view embeddings (Eq. 12), then uses $(1-S_{ij})$ as a weight on every negative term of the contrastive denominator (Eq. 16), so samples that look similar across views are not pushed apart. This replaces instance-level positives with cluster-level similarity as the unit of alignment.

What would settle it

Run the released code on Cifar100 and check whether ACC reproduces 98.53%; then, in the same setting, replace $S_{ij}$ with random weights in Eq. (16). If accuracy stays near 98%, the AsCL weighting is not driving the result; if accuracy collapses, the cross-view similarity proxy is load-bearing.

Watch

Extended reading notes

Core claim

The central claim is that the 'untrusted fusion' problem in deep multi-view clustering has two identifiable causes — noisy or redundant view information and contrastive alignment that treats only the same sample as positive — and that both are fixable in one network. TMFN segments each view-specific embedding into a token sequence, concatenates the views, and runs them through a Mamba-style selective state-space model that gates information based on the input, suppressing irrelevant view content before converting back to a fused vector. AsCL then aligns the fused representation with each view-specific representation under a contrastive loss whose denominator reweights each other sample by $(1-S_{ij})$, where $S_{ij}$ is the mean cosine similarity of view embeddings across views; samples judged to be in the same cluster are down-weighted as negatives. The result, the paper reports, is a fused representation on which k-means achieves 0.9756 ACC on Hdigit, 0.9853 on Cifar100, 0.6715 on Prokaryotic, and 0.5691 on Wiki, beating all compared deep multi-view clustering methods.

Load-bearing premise

The load-bearing premise is that the average cosine similarity $S_{ij}$ of view-specific embeddings correctly indicates whether samples $i$ and $j$ belong to the same cluster; if this proxy is wrong, AsCL will amplify the wrong pairings and the reported accuracy gains could disappear.

Editorial extensions

If this is right

  • If the reported Cifar100 result is reproducible — 98.53% ACC against CoMVC's 65.69% — then previous contrastive multi-view clustering methods were not merely slightly suboptimal but were actively misaligning representations, and cluster-aware weighting should become a standard component.
  • Ablations across the four datasets show that removing TMFN costs 3.24–9.44 ACC points and removing AsCL costs 2.13–9.92 ACC points, so the two modules contribute complementary gains rather than one doing all the work.
  • Because AsCL uses only the cross-view average cosine similarity and the contrastive loss, it requires no labels and no cluster-count supervision during training, so the method applies to any multi-view dataset with arbitrary per-view feature dimensions.
  • The reported margins imply that the reconstruction loss alone is not enough to produce clustering-friendly fused representations; the selective gate and cluster-aware contrastive term carry most of the benefit.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension the paper leaves implicit is to make $S_{ij}$ iterative: re-estimate the similarity matrix from the fused representation as training proceeds. If AsCL's benefit really comes from cluster-aware weighting, an updated $S_{ij}$ should maintain or improve accuracy; if the fixed proxy is what matters, iterative refinement should make little difference.
  • The roughly 32-point gain on Cifar100 is far outside typical incremental improvements in multi-view clustering. If it replicates independently, it would suggest that instance-level contrastive objectives were doing active harm in deep multi-view clustering, and that cluster-weighted contrastive losses could transfer to other self-supervised representation learning problems.
  • TMFN's input-dependent gate is a natural candidate for handling missing or corrupted views, since the selective state-space model can in principle learn to suppress uninformative view tokens; the paper only evaluates complete views, so this remains an untested property of the architecture.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript proposes TMCN, a deep multi-view clustering method combining autoencoder reconstruction, a 'Trusted Mamba Fusion Network' (TMFN) based on selective state-space models, and an 'Average-similarity Contrastive Learning' (AsCL) module that re-weights contrastive pairs using average cross-view cosine similarity. The authors claim state-of-the-art ACC/NMI/PUR on Hdigit, Cifar100, Prokaryotic, and Wiki, and report ablations showing both TMFN and AsCL provide large gains. The paper also includes convergence, parameter sensitivity, and t-SNE visualizations.

Significance. If validated, the use of Mamba-style selective gating for multi-view fusion and the shift from instance-level to cluster-level contrastive alignment would be a useful contribution to deep multi-view clustering. The paper promises code and provides a clear framework diagram. However, the AsCL loss appears mathematically ill-posed, the empirical state-of-the-art claim rests on single-run numbers without error bars, and at least one ablation configuration is under-specified. These issues are load-bearing, so the contribution cannot be accepted in its current form.

major comments (3)
  1. [Section II-C, Eq. (16)] The AsCL loss is not well defined as written. Since S_ii=1 for every view, the j=i term in the denominator is e^{(1-S_ii)C(...)/τ}=e^0=1; removing that positive pair would require subtracting 1, not e^{1/τ}. For τ<1, e^{1/τ} is large and can exceed the denominator sum, making the argument of the log negative and the loss undefined. The sentence after Eq. (16) additionally misstates the dependence: C is not increased by decreased S_ij; only the exponent weight (1-S_ij) changes. Because Table III credits AsCL with substantial ablation gains (e.g., 9.92 ACC on Hdigit and 2.71 on Cifar100), the SOTA claim depends on this formula. Please correct the equation or clarify if the implementation uses a different loss, and re-verify the ablation results.
  2. [Section III-B, Table II] The comparative results are reported as single values with no standard deviations, repeated runs, or seeds. On Cifar100 the reported TMCN ACC of 0.9853 versus CoMVC's 0.6569 is a 32-point margin, and the MFLVC result of 0.1342 ACC with NMI 0.0070 is far below the behavior reported in the original MFLVC paper. These numbers need to be reconciled with the evaluation protocol. Please provide mean±std over multiple runs, a description of how the deep features are obtained for each dataset, and the exact training protocol.
  3. [Section III-C, Table III and Section II-B] The 'No-TMFN' ablation is under-specified. The text says the fused representation Z is the concatenation of all view-specific representations, but Section II-B Eq. (5) already concatenates sequence vectors before the Mamba network. It is unclear whether the baseline operates on the final one-dimensional embeddings z_i^m or on the sequence-level concatenation, and whether all other components are kept identical. Without this clarification, the ablation cannot isolate the contribution of the selective mechanism.
minor comments (5)
  1. [Section II-B, Eqs. (4)-(10)] The operations rea1, rea2, rea3, and rea4 are not defined in the text or figure; please define the segmentation/reshape operation explicitly.
  2. [Section II-D, Eq. (18)] The clustering objective is written as min_{U,V} ||H - UV||_2^2 with constraints U1=1, U≥0 and called k-means, but this is not the standard k-means objective; please clarify the relationship or present the k-means update used.
  3. [Figure 2] The caption says 'convergence analysis and visualization analysis,' but only one set of curves is clearly described; please label which panel is the t-SNE visualization and which is the convergence curve.
  4. [Section III-B] The sentence 'Five state-of-the-art clustering methods' is followed by a list of five methods; the capitalization of 'Five' and the colon usage should be fixed.
  5. [Reference [25]] The page range for [25] is typeset as 'pp. 162:21 090–21 110,' which is malformed; please correct it.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the empirical SOTA claims are benchmark results, and the self-citations are background only.

full rationale

The paper's central claims are empirical: TMCN is benchmarked against five external deep multi-view clustering methods on four public datasets, and the reported ACC/NMI/PUR gains are measured, not derived from the proposed equations. The reconstruction loss (Eq. 3), TMFN selective SSM (Eqs. 4-10), and AsCL contrastive loss (Eq. 16) are defined independently of the ground-truth cluster labels; k-means (Eq. 18) is applied after the fused representation is learned. No parameter is fitted to a held-out subset of the evaluation metric and then reported as a prediction of that same metric. The self-citations (e.g., [4], [12]-[15], [17]) appear in the introduction as background on multi-view learning and hashing and are not load-bearing for the correctness or uniqueness of TMFN/AsCL. AsCL does reuse the same view-specific embeddings to form both the similarity matrix S_ij and the contrastive objective, but that is a training-signal bootstrap, not a logical reduction of the reported clustering accuracy to a fitted quantity. The apparent mathematical issue in Eq. (16) - the subtraction of e^(1/tau) in the denominator potentially making the logarithm argument non-positive - is a correctness concern, not circularity, because it does not make the result equivalent to its inputs. Thus no circular step is identified.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The paper introduces two new modules, TMFN and AsCL, but no new physical or abstract entities. Its central result depends on several free hyperparameters that are never reported and on strong domain assumptions about how cosine similarity across views relates to true cluster structure. The Mamba SSM itself is taken as a standard, pre-existing component.

free parameters (5)
  • lambda (loss weight)
    Appears in Eq. (17), L = LRec + lambda * LAsc. No value is reported in any experiment, and the balance between reconstruction and contrastive loss directly affects clustering quality.
  • tau (temperature)
    Appears in Eq. (16) inside the exponentials. No value is reported, yet the contrastive loss scale and weighting depend strongly on tau.
  • d (sequence vector dimension)
    Introduced in Eq. (4) as the dimension of the fine-grained sequence vector. Fig. 3 sweeps d on Hdigit only, and no final value is given for other datasets.
  • alpha (expansion coefficient)
    Introduced in Section II-B as d' = d * alpha for the Mamba branch. Fig. 3 sweeps alpha on Hdigit only; no final value is stated.
  • encoder/decoder/MLP hyperparameters
    The widths, depths, and activation details of the autoencoders and the several mlp blocks (mlp1 through mlp5) are not specified in the paper, though they determine the learned representations.
assumptions (4)
  • standard math The Mamba selective state-space discretization and gating equations, taken from [27], correctly model sequential feature filtering when applied to concatenated view-token sequences.
    Invoked in Section II-B, Eq. (8), where the selective SSM update h_k = A h_{k-1} + B p''_i,k is used without derivation or adaptation to the multi-view setting.
  • domain assumption The average cosine similarity S_ij over all views is a reliable proxy for cluster membership of samples i and j.
    This is the central assumption of AsCL, stated in Section II-C around Eqs. (11)-(12), and it is load-bearing: the loss denominator weighs pairs by (1 - S_ij).
  • domain assumption The Mamba selection mechanism can filter noise and redundancy in multi-view features, enabling trusted fusion.
    Claimed in Sections I and II-B without formal proof or external validation; the paper infers this from downstream clustering performance.
  • domain assumption Different views of the same sample can belong to different clusters, which motivates aligning representations by cluster rather than by instance.
    States this challenge in the Introduction and uses it to justify AsCL; it is a plausible but unverified modeling assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Trusted Mamba Contrastive Network for Multi-View Clustering." pith.science (2026). https://pith.science/paper/KOBDU2OL

@misc{pith2026241216487,
  author       = {Pith},
  title        = {Pith review of: Trusted Mamba Contrastive Network for Multi-View Clustering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KOBDU2OL}},
  note         = {Machine review of arXiv:2412.16487}
}
read the original abstract

Multi-view clustering can partition data samples into their categories by learning a consensus representation in an unsupervised way and has received more and more attention in recent years. However, there is an untrusted fusion problem. The reasons for this problem are as follows: 1) The current methods ignore the presence of noise or redundant information in the view; 2) The similarity of contrastive learning comes from the same sample rather than the same cluster in deep multi-view clustering. It causes multi-view fusion in the wrong direction. This paper proposes a novel multi-view clustering network to address this problem, termed as Trusted Mamba Contrastive Network (TMCN). Specifically, we present a new Trusted Mamba Fusion Network (TMFN), which achieves a trusted fusion of multi-view data through a selective mechanism. Moreover, we align the fused representation and the view-specific representation using the Average-similarity Contrastive Learning (AsCL) module. AsCL increases the similarity of view presentation from the same cluster, not merely from the same sample. Extensive experiments show that the proposed method achieves state-of-the-art results in deep multi-view clustering tasks. The source code is available at https://github.com/HackerHyper/TMCN.

Figures

Figures reproduced from arXiv: 2412.16487 by the authors.

Figure 1
Figure 1. Overall Framework of TMCN. The framework consists of TMFN and AsCL. TMFN segments the one-dimensional feature vector of each view into [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. The parameter analysis on Hdigit. Validity of AsCL module. According to Table III, the results of No-AsCL are lower than those of the TMCN method by 9.92, 2.71, 5.81, and 2.13 percent in ACC term. Our fused representation of multiple views is improved by the similarity of view presentation from the same cluster, rather than simply the same sample. AsCL can effectively alleviate the conflict of samples of the same cl… view at source ↗
Figure 2
Figure 2. The convergence analysis and visualization analysis on Hdigit. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 26 canonical work pages

  1. [1]

    End-to-end autonomous driving: Challenges and frontiers,

    L. Chen, P. Wu, K. Chitta, B. Jaeger, A. Geiger, and H. Li, “End-to-end autonomous driving: Challenges and frontiers,” IEEE Transactions on Pattern Analysis and Machine Intelligence , 2024

    2024

  2. [2]

    Dpnet: Dynamic poly-attention network for trustworthy multi-modal classification,

    X. Zou, C. Tang, X. Zheng, Z. Li, X. He, S. An, and X. Liu, “Dpnet: Dynamic poly-attention network for trustworthy multi-modal classification,” in Proceedings of the 31st ACM International Conference on Multimedia, 2023, pp. 3550–3559

    work page 2023

  3. [3]

    A survey on multiview clustering,

    G. Chao, S. Sun, and J. Bi, “A survey on multiview clustering,” IEEE transactions on artificial intelligence , vol. 2, no. 2, pp. 146–168, 2021

    work page 2021

  4. [4]

    Inclusivity induced adaptive graph learning for multi-view clustering,

    X. Zou, C. Tang, X. Zheng, K. Sun, W. Zhang, and D. Ding, “Inclusivity induced adaptive graph learning for multi-view clustering,” Knowledge- Based Systems, vol. 267, p. 110424, 2023

    work page 2023

  5. [5]

    Dual alignment feature embedding network for multi-omics data clustering,

    Y . Xiao, D. Yang, J. Li, X. Zou, H. Zhou, and C. Tang, “Dual alignment feature embedding network for multi-omics data clustering,”Knowledge- Based Systems, p. 112774, 2024

    work page 2024

  6. [6]

    Trustworthy multi-view clustering via alternating generative adversarial representa- tion learning and fusion,

    W. Yang, M. Wang, C. Tang, X. Zheng, X. Liu, and K. He, “Trustworthy multi-view clustering via alternating generative adversarial representa- tion learning and fusion,” Information Fusion, vol. 107, p. 102323, 2024

    work page 2024

  7. [7]

    Exploring response uncertainty in mllms: An empirical evaluation under misleading scenarios,

    Y . Dang, M. Gao, Y . Yan, X. Zou, Y . Gu, A. Liu, and X. Hu, “Exploring response uncertainty in mllms: An empirical evaluation under misleading scenarios,” arXiv preprint arXiv:2411.02708 , 2024

    arXiv 2024

  8. [8]

    Multispectral object detection via cross-modal conflict-aware learning,

    X. He, C. Tang, X. Zou, and W. Zhang, “Multispectral object detection via cross-modal conflict-aware learning,” in Proceedings of the 31st ACM International Conference on Multimedia , 2023, pp. 1465–1474

    work page 2023

Show all 33 references
  1. [9]

    Hstrans: Homogeneous substructures transformer for predicting frequencies of drug-side effects,

    K. Xu, M. Wang, X. Zou, J. Liu, A. Wei, J. Chen, and C. Tang, “Hstrans: Homogeneous substructures transformer for predicting frequencies of drug-side effects,” Neural Networks, vol. 181, p. 106779, 2025

    2025

  2. [10]

    Dai-net: Dual adaptive interaction network for coordinated medication recom- mendation,

    X. Zou, X. He, X. Zheng, W. Zhang, J. Chen, and C. Tang, “Dai-net: Dual adaptive interaction network for coordinated medication recom- mendation,” IEEE Journal of Biomedical and Health Informatics , 2024

    2024

  3. [11]

    Hierarchical attention learning for multimodal classification,

    X. Zou, C. Tang, W. Zhang, K. Sun, and L. Jiang, “Hierarchical attention learning for multimodal classification,” in 2023 IEEE International Conference on Multimedia and Expo (ICME) . IEEE, 2023, pp. 936– 941

    2023

  4. [12]

    Deep metric multi-view hashing for multimedia retrieval,

    J. Zhu, X. Ruan, Y . Cheng, Z. Huang, Y . Cui, and L. Zeng, “Deep metric multi-view hashing for multimedia retrieval,” in2023 IEEE International Conference on Multimedia and Expo (ICME) , 2023, pp. 1955–1960

    2023

  5. [13]

    Fast metric multi-view hashing for multimedia retrieval,

    J. Zhu, P. Hu, B. Li, and Y . Zhou, “Fast metric multi-view hashing for multimedia retrieval,” Information Fusion, vol. 103, p. 102130, 2024

    2024

  6. [14]

    Adaptive confidence multi-view hashing for multimedia retrieval,

    J. Zhu, Y . Cui, Z. Huang, X. Li, L. Liu, L. Zeng, and L.-R. Dai, “Adaptive confidence multi-view hashing for multimedia retrieval,” in ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2024, pp. 7900–7904

    2024

  7. [15]

    Central similarity multi-view hashing for multimedia retrieval,

    J. Zhu, W. Cheng, Y . Cui, C. Tang, Y . Dai, Y . Li, and L. Zeng, “Central similarity multi-view hashing for multimedia retrieval,” in Web and Big Data. Springer Nature Singapore, 2024, pp. 486–500

    2024

  8. [16]

    Look twice before you answer: Memory-space visual retracing for hallucination mitigation in multimodal large language models,

    X. Zou, Y . Wang, Y . Yan, S. Huang, K. Zheng, J. Chen, C. Tang, and X. Hu, “Look twice before you answer: Memory-space visual retracing for hallucination mitigation in multimodal large language models,” arXiv preprint arXiv:2410.03577, 2024

    arXiv

  9. [17]

    Boosted curriculum multi-view hashing for multimedia retrieval,

    J. Zhu, Z. Huang, L. Liu, C. Tang, and L.-R. Dai, “Boosted curriculum multi-view hashing for multimedia retrieval,” IEEE Signal Processing Letters, vol. 31, pp. 2065–2069, 2024

    2024

  10. [18]

    Deep multiple auto- encoder-based multi-view clustering,

    G. Du, L. Zhou, Y . Yang, K. L ¨u, and L. Wang, “Deep multiple auto- encoder-based multi-view clustering,” Data Science and Engineering , vol. 6, no. 3, pp. 323–338, 2021

    2021

  11. [19]

    Deep multimodal subspace clustering networks,

    M. Abavisani and V . M. Patel, “Deep multimodal subspace clustering networks,” IEEE Journal of Selected Topics in Signal Processing , vol. 12, no. 6, pp. 1601–1614, 2018

    2018

  12. [20]

    End-to-end adversarial-attention network for multi-modal clustering,

    R. Zhou and Y .-D. Shen, “End-to-end adversarial-attention network for multi-modal clustering,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2020, pp. 14 619–14 628

    2020

  13. [21]

    Deep embedded multi- view clustering with collaborative training,

    J. Xu, Y . Ren, G. Li, L. Pan, C. Zhu, and Z. Xu, “Deep embedded multi- view clustering with collaborative training,” Information Sciences , vol. 573, pp. 279–290, 2021

    2021

  14. [22]

    Reconsider- ing representation alignment for multi-view clustering,

    D. J. Trosten, S. Lokse, R. Jenssen, and M. Kampffmeyer, “Reconsider- ing representation alignment for multi-view clustering,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2021, pp. 1255–1265

    2021

  15. [23]

    Multi-level feature learning for contrastive multi-view clustering,

    J. Xu, H. Tang, Y . Ren, L. Peng, X. Zhu, and L. He, “Multi-level feature learning for contrastive multi-view clustering,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 16 051–16 060

    2022

  16. [24]

    Approximating the kullback leibler di- vergence between gaussian mixture models,

    J. R. Hershey and P. A. Olsen, “Approximating the kullback leibler di- vergence between gaussian mixture models,” in 2007 IEEE International Conference on Acoustics, Speech and Signal Processing-ICASSP’07 , vol. 4. IEEE, 2007, pp. IV–317

    2007

  17. [25]

    Deep safe incomplete multi-view clustering: Theorem and algorithm,

    H. Tang and Y . Liu, “Deep safe incomplete multi-view clustering: Theorem and algorithm,” in Proceedings of the 39th International Conference on Machine Learning , 2022, pp. 162:21 090–21 110

    2022

  18. [26]

    Deep incomplete multi-view clustering via mining cluster complementarity,

    J. Xu, C. Li, Y . Ren, L. Peng, Y . Mo, X. Shi, and X. Zhu, “Deep incomplete multi-view clustering via mining cluster complementarity,” in Thirty-Six AAAI conference on artificial intelligence , 2022, pp. 8761– 8769

    2022

  19. [27]

    Mamba: Linear-time sequence modeling with selective state spaces,

    A. Gu and T. Dao, “Mamba: Linear-time sequence modeling with selective state spaces,” CoRR, vol. abs/2312.00752, 2023

    2023 arXiv

  20. [28]

    Self- supervised video hashing with hierarchical binary auto-encoder,

    J. Song, H. Zhang, X. Li, L. Gao, M. Wang, and R. Hong, “Self- supervised video hashing with hierarchical binary auto-encoder,” IEEE Transactions on Image Processing, vol. 27, no. 7, pp. 3210–3221, 2018

    2018

  21. [29]

    Reducing the dimensionality of data with neural networks,

    G. E. Hinton and R. R. Salakhutdinov, “Reducing the dimensionality of data with neural networks,” science, vol. 313, no. 5786, pp. 504–507, 2006

    2006

  22. [30]

    Gcfagg: Global and cross-view feature aggregation for multi-view clustering,

    W. Yan, Y . Zhang, C. Lv, C. Tang, G. Yue, L. Liao, and W. Lin, “Gcfagg: Global and cross-view feature aggregation for multi-view clustering,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2023, pp. 19 863–19 872

    2023

  23. [31]

    D. J. MacKay, D. J. Mac Kay et al., Information theory, inference and learning algorithms. Cambridge university press, 2003

    2003

  24. [32]

    K-means clustering is matrix factorization,

    C. Bauckhage, “K-means clustering is matrix factorization,” arXiv preprint arXiv:1512.07548, 2015

    2015 arXiv

  25. [33]

    Visualizing data using t-sne

    L. Van der Maaten and G. Hinton, “Visualizing data using t-sne.” Journal of machine learning research , vol. 9, no. 11, 2008

    2008

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.