Pith. sign in

REVIEW 4 major objections 6 minor 43 references

Information-Theoretic Generative Clustering of Documents

T0 review · 4 major / 6 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 Clustering by LLM-generated text outdoes every baseline

desk verdict Novel LLM-distribution clustering with correct math, but the SOTA claim rests on alpha=0.25 selected on test labels; at alpha=1 the gains mostly vanish. read the letter →

arxiv 2412.13534 v1 pith:O5EJMYI6 submitted 2024-12-18 cs.LG cs.CLcs.IRcs.ITmath.IT

Xin Du , Kumiko Tanaka-Ishii This is my paper · ORCID
classification cs.LGcs.CLcs.IRcs.ITmath.IT
keywords generativeclusteringdocumentlargelanguagemodelsKLdivergenceimportancesamplingretrievalregularizedLLM-generatedtexts
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

Instead of clustering documents directly, generative clustering (GC) turns each document $x$ into a probability distribution $p(Y|x)$ over texts that a large language model might generate from $x$, and measures how far two documents are by the KL divergence between these distributions. The paper's central claim is that this information-theoretic reformulation, together with a regularized importance-sampling algorithm, clusters documents more accurately than any previous method, often by a large margin, on R2, R5, AG News, and Yahoo! Answers. The same construction, applied hierarchically to build document indexes for generative retrieval, improves Recall@1 by up to 36 percent over a strong retrieval baseline on MS Marco Lite. A sympathetic reader would care because the result suggests that the unsaid knowledge an LLM supplies about a text is genuinely useful for recovering cluster structure, and because the formulation replaces heuristic vector-plus-Euclidean-distance pipelines with a principled distributional objective.

What carries the argument

The central object is the per-document distribution $p(Y|x)$ over the infinite set $Y$ of all possible word sequences, computed by multiplying word-level probabilities from an LLM. The load-bearing identity is the importance-sampling rewrite of the KL divergence: $\mathbb{E}_{Y\sim p(Y|x)}\log\frac{p(Y|x)}{p(Y|k)} = \mathbb{E}_{Y\sim\phi}\left[\frac{p(Y|x)}{\phi(Y)}\log\frac{p(Y|x)}{p(Y|k)}\right]$, which turns an intractable divergence into a finite-sample estimator over a shared set of generated texts. The proposal $\phi$ is set to the second-moment-minimizing distribution (Eq. 7), the estimator uses regularized weights $W_{ij}=(p(y_j|x_i)/\phi(y_j))^\alpha$ with $\alpha=0.25$, and cluster centroids are normalized sums of those weights, making the whole procedure a Bregman hard-clustering iteration.

What would settle it

On any held-out corpus, compute GC's ACC, NMI, and ARI with $\alpha=0.25$ and with $\alpha$ tuned on a disjoint validation split; if the tuned value differs substantially or the fixed 0.25 version loses to the best SBERT embedding baseline, the claim that 0.25 is a universally robust setting collapses.

Watch

Extended reading notes

Core claim

Generative clustering defines a document not by a fixed vector but by the conditional distribution $p(Y=y|x)$ over the infinite set of all word sequences, with each word probability computed autoregressively by an LLM. The distortion between document $x$ and cluster centroid $k$ is the KL divergence $\mathrm{KL}(p(Y|x)\|p(Y|k))$; because $Y$ is infinite, this is estimated by regularized importance sampling from a proposal $\phi$ chosen to minimize the second moment of the importance weights, yielding the estimator in Eq. (5) with exponent $\alpha=0.25$. The algorithm alternates between assigning each document to its closest centroid and updating each centroid to the normalized mean of the importance-weight rows in its cluster, a procedure proven to converge to a local minimum. On four datasets, this method is reported to outperform all compared baselines on accuracy, NMI, and ARI, with the largest gap on R2 (96.1% accuracy and 77.8 NMI versus 65.6 NMI for the best SBERT baseline). The paper also reports that applying the same KL-based hierarchical clustering to build prefix-code indexes raises retrieval accuracy in generative document retrieval beyond two prior indexing methods.

Load-bearing premise

The method's superiority rests on the regularization exponent $\alpha=0.25$, which was selected by measuring clustering performance on the same four datasets that are later reported as results; the paper gives no derived or held-out justification that this value generalizes to new collections.

Editorial extensions

If this is right

  • On the four tested datasets, GC beats every compared clustering baseline on ACC, NMI, and ARI; on R2 it cuts the error rate from 8.0% to 3.9%.
  • Using $\alpha=0.25$ instead of the unbiased $\alpha=1$ improves NMI on R2 from 25.8 to 77.8, showing that variance reduction, not unbiasedness, is what makes the estimator recover cluster structure.
  • Performance is stable once the number of generated texts $J$ is at least 50 on small datasets and 384 on Yahoo! Answers, below the 768-dimensional BERT embedding.
  • When $K$ is misspecified, GC still outperforms the best SBERT baseline across $K$ from 2 to 20 on R5.
  • Applied hierarchically to prefix-code indexing, GC raises Recall@1 on MS Marco Lite by 36% over BMI and MRR@100 by 25%.
  • The proposal distribution remains effective when localized to sub-clusters during hierarchical clustering, and removing that localization reduces retrieval accuracy.

Reading between the lines

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

  • The $\alpha=0.25$ choice was made by evaluating on the same datasets reported; a fairer test would tune $\alpha$ on a validation split, and the paper does not show that 0.25 transfers to unrelated corpora.
  • If the gains come from the LLM's ability to explicate implicit content, then using a more capable or domain-adapted generator should improve clustering further; likewise, datasets whose documents are not query-like may show smaller gains, since the doc2query model is specialized for generating queries.
  • The same regularized-importance-sampling trick could be applied to other infinite-space divergence problems, such as clustering images or code by their generation probabilities, whenever a generative model supplies exact conditional probabilities.
  • A direct extension would be to use the generated-text distribution as a feature for supervised tasks, or to check whether the importance-weight matrix $W$ itself provides interpretable cluster labels through its highest-weight texts.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes 'generative clustering' (GC), in which each document is represented by the conditional distribution p(Y|x) over texts generated by a large language model, and cluster centroids are distributions over the same space. The distortion between a document and a centroid is the KL divergence, estimated by regularized importance sampling with a proposal distribution p(Y), a clipping step, and an exponent alpha that controls the bias-variance tradeoff. The authors provide an alternating assignment/centroid-update algorithm, prove convergence and centroid optimality for the estimated objective, and report experiments on R2, R5, AG News, and Yahoo! Answers claiming state-of-the-art clustering performance. They also apply the method to hierarchical indexing for generative document retrieval and report improved retrieval accuracy on MS Marco Lite and NQ320K.

Significance. If the central claim were established for a genuinely unsupervised setting, this would be a valuable contribution: it gives a principled information-theoretic formulation of LLM-based document clustering, introduces an importance-sampling estimator with an explicit bias-variance knob, and demonstrates a credible application to generative retrieval. The algorithmic ideas are novel relative to the cited k-means-on-embeddings baselines, the code is released, and the appendix contains formal propositions for convergence and centroid optimality. However, the paper's headline 'state-of-the-art, outperforming any previous clustering method' claim is currently tied to a hyperparameter value (alpha=0.25) that is selected using ground-truth labels on the same four evaluation datasets, and the unbiased version (alpha=1) performs far worse, often below standard SBERT baselines. That weakness is load-bearing for the main empirical claim.

major comments (4)
  1. [§6.2, §6.3, Figure 1(a-d), Table 2] The value alpha=0.25 is selected by evaluating ACC, NMI, and ARI on the same labeled datasets that are later reported as the evaluation results. The ablations in Table 2 show this choice is decisive: with alpha=1, R2 NMI drops from 77.8 to 25.8, and the other datasets show large losses as well, often falling below the best SBERT baselines. Since clustering is presented as an unsupervised task, tuning the core hyperparameter on the test labels invalidates the claim that GC 'outperforms any previous clustering method' in a genuinely unsupervised setting. The authors need a held-out validation split, a nested cross-validation procedure, or a theoretical derivation that selects alpha=0.25 without label access. The qualitative bias-variance argument in Section 8 does not supply such a derivation.
  2. [§5.2, Eq. (5), Propositions 2-3] The paper proves optimality of the centroid update for the estimated objective and derives the proposal that minimizes the second moment of the importance weights, but it never shows that the biased estimator with alpha<1 preserves the cluster structure of the true KL divergence, i.e., that the ordering of pairwise distances d(x,k) is approximately maintained. This is not merely a technical gap: the alpha=1 ablation performs so much worse that the claimed superiority is attributable to the particular bias induced by alpha=0.25 rather than to the information-theoretic objective itself. A formal or systematic empirical statement about when the biased estimator preserves cluster structure is needed to support the method's name and its claimed advantage.
  3. [§2.1, §6.3, Table 2] The abstract and Section 6.3 claim that GC outperforms 'any previous clustering method', but Table 2 contains no recent LLM-based clustering baselines, even though Viswanathan et al. (2023) and Zhang, Wang, and Shang (2023) are cited in Related Work. The comparison set consists of older k-means, DEC/IDEC/STC/DFTC, and embedding methods. The state-of-the-art claim is therefore not supported by the experiments as presented. The authors should either benchmark against those recent methods or explicitly qualify the claim to the compared baselines.
  4. [§7, Table 3] The retrieval experiment repeats the same alpha-selection issue: alpha=0.25 gives the best retrieval numbers, and the ablations show alpha=1 is substantially worse (e.g., MS Marco Lite Rec@1 drops from 32.41 to 24.43). The paper states that alpha=0.25 is a 'robust choice' without a validation procedure for the retrieval task. If alpha is selected using the clustering labels on the four datasets and then transferred to retrieval, the retrieval result is not an independent confirmation; it is a second instance of the same fitted hyperparameter.
minor comments (6)
  1. [§3, first paragraph] The phrase 'The condition probability' should be 'The conditional probability'.
  2. [§4, Eq. (4)] Equation (4) has mismatched parentheses: the norm expression 'E[p(Y |f (x))∥2' is missing a closing bracket and uses a nonstandard norm notation; it should read '∥E[p(Y |x)] − E[p(Y |f(x))]∥²'.
  3. [§5.1, Algorithm 1 line 1 and §5.3] Algorithm 1 says Y is generated by i.i.d. sampling 'from the language model', but the actual procedure in Section 5.3 first samples a document uniformly and then generates a text from it. The algorithm should state this two-step sampling procedure explicitly.
  4. [Appendix A.2, proof of Proposition 2] The proof ends with 'This completes the proof of Proposition 5.4', but the proposition is numbered Proposition 2 in the appendix. The cross-reference should be corrected.
  5. [§5.2, Eq. (5)] The notation 'yj ∼ ϕ(Y)' under the summation is confusing; it should be 'yj ∼ ϕ' or 'yj drawn from ϕ'.
  6. [Table 2] Standard deviations are reported only for the GC row. Reporting them for the strongest baselines would help the reader judge whether margins of a few points (e.g., Yahoo! Answers ACC 60.7 vs. 58.2 for all-minilm-l12-v2) are meaningful.

Circularity Check

1 steps flagged · score 6.0 of 10

The state-of-the-art claim is reported for α=0.25, which was selected by sweeping the same ACC/NMI/ARI metrics on the same four datasets; at α=1 the advantage largely collapses.

  1. fitted input called prediction [Section 5.2 (α setting); Section 6.3 and Figure 1(a–d) (α sweep on ACC/NMI/ARI); Table 2 (ablated α=1 row)]
    "In this paper, α was set to 0.25 for all experiments. ... α = 0.25 consistently yielded the best performance across all datasets. ... The largest performance loss occurred with α = 1.00 instead of α = 0.25, an interesting outcome that we discuss in Section 8. ... This explains why reducing α from 1 to 0.25 significantly improved the NMI on the R2 dataset from 25.8% to 77.8% (Table 2)."

    The headline result (GC outperforms all baselines) is reported for α=0.25, but Figures 1(a–d) sweep α on exactly the four datasets and three metrics (ACC, NMI, ARI) that Table 2 reports, and the text states α=0.25 consistently yielded the best performance across all datasets. Thus the reported GC row is, by construction, the argmax of the evaluation metrics over α on the same test sets; no held-out validation split or derivation fixes 0.25. The ablated α=1 (the unbiased KL estimator) row shows the advantage largely vanishes (e.g., R2 NMI drops from 77.8 to 25.8), so the claimed state of the art is carried by the test-selected regularization, not by the derived information-theoretic objective. This is the fitted-input-called-prediction pattern.

full rationale

The formal derivation is not circular: the distortion in Eqs. (2)–(3) is Bregman hard clustering; Proposition 2 proves the centroid update in Eq. (10); Proposition 3 derives the proposal ϕ in Eq. (7) from a second-moment minimization. The doc2query model, BERT/SBERT baselines, and NQ320K retrieval benchmark are external, and no load-bearing self-citation chain is present; the authors' own MS Marco Lite dataset is supplemented by NQ320K. The single significant circularity is empirical: α=0.25 was selected using the target ACC/NMI/ARI metrics on the same four datasets that are then reported as results, and Table 2's α=1 ablation shows the advantage mostly disappears without that fitted value. The score of 6 reflects partial circularity: the central empirical claim reduces to the test-selected hyperparameter, while the algorithmic derivation retains independent content.

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

The method introduces no new physical or conceptual entities. It relies on the LLM as a fixed probabilistic model, plus three hand-chosen hyperparameters (alpha, J, C), with alpha being the most consequential and tuned on the evaluation data.

free parameters (4)
  • alpha (regularization exponent) = 0.25
    Tuned to maximize ACC/NMI/ARI on the four evaluation datasets; Figures 1(a-d) show it consistently best, but the unbiased choice alpha=1 gives much worse results.
  • J (number of generated texts) = 1024 for clustering, 4096 for retrieval
    Chosen by hand; performance converges for J above 50 on small datasets, and the specific value is not derived.
  • C (centroid probability mass) = 1
    Set to 1 for all clusters; the paper shows distance ordering is invariant to C, so it is a benign but arbitrary choice.
  • Clipping threshold = 5 sigma
    Fixed heuristic in Eq. (11); not tuned in the paper.
assumptions (4)
  • domain assumption Documents X are i.i.d. samples from a prior p(X), so the marginal p(Y) is the average of p(Y|x) over x.
    Used in Section 5.3 to justify sampling y via a random document and estimating phi(y) by averaging p(y|x)^{2alpha}.
  • domain assumption The LLM's probability p(y|x) faithfully reflects document content, and KL divergence between such distributions is the correct clustering objective.
    Central modeling choice in Section 3; no independent evidence that this objective is superior to standard embedding distances.
  • ad hoc to paper The biased importance-sampling estimator with alpha<1 preserves the cluster structure of the true KL divergence.
    The paper observes empirically that alpha=0.25 works best, but there is no proof that the bias preserves ordering; Section 8 discusses a bias-variance tradeoff qualitatively.
  • standard math Standard Bregman hard clustering convergence and Lagrange multiplier optimality.
    Used in Appendix A.1 and A.2; these are textbook results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Information-Theoretic Generative Clustering of Documents." pith.science (2026). https://pith.science/paper/O5EJMYI6

@misc{pith2026241213534,
  author       = {Pith},
  title        = {Pith review of: Information-Theoretic Generative Clustering of Documents},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O5EJMYI6}},
  note         = {Machine review of arXiv:2412.13534}
}
abstract

We present {\em generative clustering} (GC) for clustering a set of documents, $\mathrm{X}$, by using texts $\mathrm{Y}$ generated by large language models (LLMs) instead of by clustering the original documents $\mathrm{X}$. Because LLMs provide probability distributions, the similarity between two documents can be rigorously defined in an information-theoretic manner by the KL divergence. We also propose a natural, novel clustering algorithm by using importance sampling. We show that GC achieves the state-of-the-art performance, outperforming any previous clustering method often by a large margin. Furthermore, we show an application to generative document retrieval in which documents are indexed via hierarchical clustering and our method improves the retrieval accuracy.

Figures

Figures reproduced from arXiv: 2412.13534 by the authors.

Figure 1
Figure 1. Performance of GC on the four datasets (columns) with varying (a-d) [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 28 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Aouali, I.; Brunel, V.-E.; Rohde, D.; and Korba, A. 2024. Unified PAC-Bayesian Study of Pessimism for Offline Policy Learning with Regularized Importance Sampling. arXiv preprint arXiv:2406.03434

    work page Pith review arXiv 2024

  4. [4]

    Arthur, D.; and Vassilvitskii, S. 2006. k-means++: The advantages of careful seeding. Technical report, Stanford

    work page 2006

  5. [5]

    S.; Ghosh, J.; and Lafferty, J

    Banerjee, A.; Merugu, S.; Dhillon, I. S.; Ghosh, J.; and Lafferty, J. 2005. Clustering with Bregman divergences. Journal of machine learning research, 6(10)

    work page 2005

  6. [6]

    M.; Ng, A

    Blei, D. M.; Ng, A. Y.; and Jordan, M. I. 2003. Latent dirichlet allocation. Journal of machine Learning research, 3(Jan): 993--1022

    2003

  7. [7]

    Devlin, J.; Chang, M.-W.; Lee, K.; and Toutanova, K. 2019. BERT : Pre-training of Deep Bidirectional Transformers for Language Understanding. In Burstein, J.; Doran, C.; and Solorio, T., eds., Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and ...

    2019

  8. [8]

    S.; Mallela, S.; and Modha, D

    Dhillon, I. S.; Mallela, S.; and Modha, D. S. 2003. Information-theoretic co-clustering. In Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining, 89--98

    work page 2003

Show all 43 references
  1. [9]

    Du, X.; Xiu, L.; and Tanaka-Ishii, K. 2024. Bottleneck-Minimal Indexing for Generative Document Retrieval. In Forty-first International Conference on Machine Learning

    2024

  2. [10]

    Guan, R.; Zhang, H.; Liang, Y.; Giunchiglia, F.; Huang, L.; and Feng, X. 2022. Deep feature-based text clustering and its explanation. IEEE Transactions on Knowledge and Data Engineering, 34(8): 3669--3680

    2022

  3. [11]

    Gulli, A. 2005. AG news homepage

    2005

  4. [12]

    Guo, X.; Gao, L.; Liu, X.; and Yin, J. 2017. Improved deep embedded clustering with local structure preservation. In Ijcai, volume 17, 1753--1759

    2017

  5. [13]

    Hesterberg, T. 1995. Weighted average importance sampling and defensive mixture distributions. Technometrics, 37(2): 185--194

    1995

  6. [14]

    Hubert, L.; and Arabie, P. 1985. Comparing partitions. Journal of classification, 2: 193--218

    1985

  7. [15]

    Kloek, T.; and Van Dijk, H. K. 1978. Bayesian estimates of equation system parameters: an application of integration by Monte Carlo. Econometrica: Journal of the Econometric Society, 1--19

    1978

  8. [16]

    S.; Reid, M.; Matsuo, Y.; and Iwasawa, Y

    Kojima, T.; Gu, S. S.; Reid, M.; Matsuo, Y.; and Iwasawa, Y. 2022. Large language models are zero-shot reasoners. Advances in neural information processing systems, 35: 22199--22213

    2022

  9. [17]

    Korba, A.; and Portier, F. 2022. Adaptive importance sampling meets mirror descent: a bias-variance tradeoff. In International Conference on Artificial Intelligence and Statistics, 11503--11527. PMLR

    2022

  10. [18]

    Kwiatkowski, T.; Palomaki, J.; Redfield, O.; Collins, M.; Parikh, A.; Alberti, C.; Epstein, D.; Polosukhin, I.; Devlin, J.; Lee, K.; et al. 2019. Natural questions: a benchmark for question answering research. Transactions of the Association for Computational Linguistics, 7: 453--466

    2019

  11. [19]

    S.; and Dean, J

    Mikolov, T.; Sutskever, I.; Chen, K.; Corrado, G. S.; and Dean, J. 2013. Distributed representations of words and phrases and their compositionality. Advances in neural information processing systems, 26

    2013

  12. [20]

    J.; and Jordan, M

    Nguyen, X.; Wainwright, M. J.; and Jordan, M. I. 2010. Estimating divergence functionals and the likelihood ratio by convex risk minimization. IEEE Transactions on Information Theory, 56(11): 5847--5861

    2010

  13. [21]

    Nogueira, R.; Lin, J.; and Epistemic, A. 2019. From doc2query to docTTTTTquery. Online preprint, 6: 2

    2019

  14. [22]

    OpenAI. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774

    2023 arXiv

  15. [23]

    Owen, A.; and Zhou, Y. 2000. Safe and effective importance sampling. Journal of the American Statistical Association, 95(449): 135--143

    2000

  16. [24]

    P \'e rez-Cruz, F. 2008. Kullback-Leibler divergence estimation of continuous distributions. In 2008 IEEE international symposium on information theory, 1666--1670. IEEE

    2008

  17. [25]

    E.; Neumann, M.; Iyyer, M.; Gardner, M.; Clark, C.; Lee, K.; and Zettlemoyer, L

    Peters, M. E.; Neumann, M.; Iyyer, M.; Gardner, M.; Clark, C.; Lee, K.; and Zettlemoyer, L. 2018. Deep Contextualized Word Representations. In Walker, M.; Ji, H.; and Stent, A., eds., Proceedings of the 2018 Conference of the North A merican Chapter of the Association for Comp...

    2018

  18. [26]

    Raffel, C.; Shazeer, N.; Roberts, A.; Lee, K.; Narang, S.; Matena, M.; Zhou, Y.; Li, W.; and Liu, P. J. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. The Journal of Machine Learning Research, 21(1): 5485--5551

    2020

  19. [27]

    Reimers, N.; and Gurevych, I. 2019. Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics

    2019

  20. [28]

    Slonim, N.; and Tishby, N. 2000. Document clustering using word clusters via the information bottleneck method. In Proceedings of the 23rd annual international ACM SIGIR conference on Research and development in information retrieval, 208--215

    2000

  21. [29]

    Subakti, A.; Murfi, H.; and Hariadi, N. 2022. The performance of BERT as data representation of text clustering. Journal of big Data, 9(1): 15

    2022

  22. [30]

    W.; and Metzler, D

    Tay, Y.; Tran, V.; Dehghani, M.; Ni, J.; Bahri, D.; Mehta, H.; Qin, Z.; Hui, K.; Zhao, Z.; Gupta, J.; Schuster, T.; Cohen, W. W.; and Metzler, D. 2022. Transformer Memory as a Differentiable Search Index. In Koyejo, S.; Mohamed, S.; Agarwal, A.; Belgrave, D.; Cho, K.; and Oh, ...

    2022

  23. [31]

    X.; Epps, J.; and Bailey, J

    Vinh, N. X.; Epps, J.; and Bailey, J. 2010. Information Theoretic Measures for Clusterings Comparison: Variants, Properties, Normalization and Correction for Chance. Journal of Machine Learning Research, 11(95): 2837--2854

    2010

  24. [32]

    Viswanathan, V.; Gashteovski, K.; Lawrence, C.; Wu, T.; and Neubig, G. 2023. Large language models enable few-shot clustering. arXiv preprint arXiv:2307.00524

    2023 arXiv

  25. [33]

    R.; and Verd \'u , S

    Wang, Q.; Kulkarni, S. R.; and Verd \'u , S. 2009. Divergence estimation for multidimensional densities via k -Nearest-Neighbor distances. IEEE Transactions on Information Theory, 55(5): 2392--2405

    2009

  26. [34]

    Wang, Y.; Hou, Y.; Wang, H.; Miao, Z.; Wu, S.; Sun, H.; Chen, Q.; Xia, Y.; Chi, C.; Zhao, G.; Liu, Z.; Xie, X.; Sun, H.; Deng, W.; Zhang, Q.; and Yang, M. 2022. A Neural Corpus Indexer for Document Retrieval. In Oh, A. H.; Agarwal, A.; Belgrave, D.; and Cho, K., eds., Advances...

    2022

  27. [35]

    G.; Xing, E.; and Hu, Z

    Wu, Y.; Zhou, P.; Wilson, A. G.; Xing, E.; and Hu, Z. 2020. Improving gan training with probability ratio clipping and sample reweighting. Advances in Neural Information Processing Systems, 33: 5729--5740

    2020

  28. [36]

    Xie, J.; Girshick, R.; and Farhadi, A. 2016. Unsupervised deep embedding for clustering analysis. In International conference on machine learning, 478--487. PMLR

    2016

  29. [37]

    Xu, J.; Xu, B.; Wang, P.; Zheng, S.; Tian, G.; and Zhao, J. 2017. Self-taught convolutional neural networks for short text clustering. Neural Networks, 88: 22--31

    2017

  30. [38]

    Yang, X.; Yan, J.; Cheng, Y.; and Zhang, Y. 2022. Learning deep generative clustering via mutual information maximization. IEEE Transactions on Neural Networks and Learning Systems, 34(9): 6263--6275

    2022

  31. [39]

    Yin, J.; and Wang, J. 2014. A dirichlet multinomial mixture model-based approach for short text clustering. In Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining, 233--242

    2014

  32. [40]

    Zhang, X.; Zhao, J.; and LeCun, Y. 2015. Character-level convolutional networks for text classification. Advances in neural information processing systems, 28

    2015

  33. [41]

    Zhang, Y.; Wang, Z.; and Shang, J. 2023. Clusterllm: Large language models as a guide for text clustering. arXiv preprint arXiv:2305.14871

    2023 arXiv

  34. [42]

    Zhong, S.; and Ghosh, J. 2005. Generative model-based document clustering: a comparative study. Knowledge and Information Systems, 8: 374--384

    2005

  35. [43]

    Zhou, S.; Xu, H.; Zheng, Z.; Chen, J.; Bu, J.; Wu, J.; Wang, X.; Zhu, W.; Ester, M.; et al. 2022. A comprehensive survey on deep clustering: Taxonomy, challenges, and future directions. arXiv preprint arXiv:2206.07579

    2022 arXiv

Pith tools

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