Pith. sign in

REVIEW 3 major objections 6 minor 45 references

Embedding Style Beyond Topics: Analyzing Dispersion Effects Across Different Language Models

T0 review · 3 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Writing style measurably changes the spread of text embeddings in language models, but topic variation spreads them more.

desk verdict A carefully built corpus for separating style and topic, but the style-specific conclusion is unidentifiable because every style comparison confounds authorship with GPT-4o generation. read the letter →

arxiv 2501.00828 v1 pith:ZE7KS4U6 submitted 2025-01-01 cs.CL cs.AI

Benjamin Icard , Evangelia Zve , Lila Sainero , Alice Breton , Jean-Gabriel Ganascia This is my paper · ORCID
classification cs.CLcs.AI
keywords embeddingdispersionwritingstyletopicmodelinglanguagemodelsUMAPstylometrytextembeddingsmultilingual
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

The paper asks whether writing style leaves a measurable trace in the geometry of embedding vectors, and whether topic does so more strongly. It builds a bilingual corpus of literary texts in which topic and style are swapped independently: one source has one topic repeated in many styles, another has many topics in one style, and language-model rewrites create the crossed combinations. Measuring mean distance of each text from its class centroid in UMAP projections, it finds that both topic variation and style variation increase embedding dispersion, with topic variation producing larger increases. The authors interpret this as evidence that style is encoded in embedding spaces, but secondary to topic.

What carries the argument

The central object is the QUENEAU-FENEON corpus, a four-cell design in which topic and style vary independently. Each language has 73 texts in each of: same topic with varied styles, varied topics with the same style, and two language-model-generated crossed classes. The central metric is the mean Euclidean distance from each text embedding to its class centroid in UMAP-reduced space, averaged over 30 random seeds. The hypotheses are inequalities between these means: topic variation should increase dispersion, style variation should increase dispersion, and the topic gap should exceed the style gap. This metric converts the abstract question of whether embeddings encode style into a directly testable ordering of numbers.

What would settle it

One concrete test is to generate rewrites that preserve both topic and style while changing only surface wording, then measure whether mean centroid distance changes as much as it does in the style- or topic-change conditions; if it does, dispersion is responding to text generation itself rather than to style or topic. A second test is to run the same dispersion comparisons on a human-authored corpus with topic and style varied independently, without any machine rewrites.

Watch

Extended reading notes

Core claim

The central claim is stated in the conclusion: writing style influences embedding dispersion, and topic variation has a stronger effect. Concretely, the paper predicts and observes the order $$\bar d_{\mathrm{FENEON\_GEN}} > \bar d_{\mathrm{FENEON\_REF}} > \bar d_{\mathrm{QUENEAU\_REF}} > \bar d_{\mathrm{QUENEAU\_GEN}}$$ in mean centroid distance, with pairwise differences significant at the .01 level across most models and both languages. The local hypotheses (T) and (S) attribute the first two gaps to topic heterogeneity and the style gap to style heterogeneity, while the global hypothesis (T-S) attributes the FENEON_REF-to-QUENEAU_REF gap to topic dominating style. Attempted interpretability links dispersion to readability and complexity indexes, function words, and punctuation.

Load-bearing premise

The entire attribution rests on the assumption that the rewriting step changes only the intended dimension, style or topic, and does not systematically alter embedding geometry for other reasons.

Editorial extensions

If this is right

  • Stylistic differences among authors are detectable in embedding geometry, so style-sensitive tasks such as authorship verification should treat embedding similarity as carrying style information.
  • Topic must be controlled before interpreting dispersion as a style signal in any embedding-based analysis.
  • Models vary in how strongly they encode style; applications that rely on style, such as style transfer evaluation, should not assume all embedding models respond equally.
  • Readability and complexity indexes, function words, and punctuation are the surface features most associated with the style-driven dispersion.
  • Translation can dampen the stylistic signal, so multilingual style comparisons need to check that feature frequencies survive translation.

Reading between the lines

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

  • If this dispersion signature is stable, embedding dispersion could serve as an unsupervised proxy for detecting whether a text has been stylistically rewritten, without labelled training data.
  • The same four-cell design could be applied to typologically distant languages; the relative strength of the topic effect may shift if style is carried by different surface features.
  • A direct extension would replace the mean-centroid-distance scalar with shape descriptors, such as variance along principal axes, to separate style- and topic-specific directions rather than collapsing them into one number.
  • The English rewriting prompt imposed a word limit not present in the French prompt, so an immediate replication with identical prompt constraints across languages would test whether the observed French-English gap is linguistic or procedural.
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

3 major / 6 minor

Summary. The paper introduces the QUENEAU-FENEON corpus, built from two French literary works and their English translations, with two human reference classes (QUENEAU_REF: 73 same-topic, multi-style texts; FENEON_REF: 73 multi-topic, same-style texts) and two GPT-4o-generated classes (QUENEAU_GEN: same topic, uniform Fénéon style; FENEON_GEN: varied topics, varied Queneau styles). For twelve embedding models and two languages, the authors measure within-class dispersion as the mean Euclidean distance to the class centroid in UMAP-reduced embedding space, and test hypotheses that topic variation and style variation increase dispersion, with topic having the larger effect. They also correlate per-text dispersion differences with differences in eight stylistic feature groups. The paper concludes that writing style influences embedding dispersion, though topic variation has a stronger effect.

Significance. If the central claim were established, the paper would offer a compact, reproducible framework for comparing how different embedding models encode style versus topic: the corpus construction is transparent, the experiments span twelve models and two languages, and code and data are promised in a GitHub repository. The one clean human-human comparison, (T-S) with dbar(FENEON_REF) > dbar(QUENEAU_REF), is a reasonably supported demonstration that topic variation is associated with larger dispersion than style variation in these two literary corpora. However, the paper's stronger claim that style alone drives dispersion is not identifiable from the reported comparisons, because the style hypotheses are tested on human-versus-machine comparisons that also change authorship, generation artifacts, and, for English, text length.

major comments (3)
  1. [§4.2, hypotheses (S') and (S'')] The two style hypotheses are tested on comparisons that vary authorship and generation procedure alongside style. (S') compares QUENEAU_REF (73 human-written Queneau exercises) with QUENEAU_GEN (GPT-4o rewrites of the same stories in a single Fénéon style); the English prompt additionally imposes a strict length constraint ('strictly less than 30 words and using only 1 to 3 sentences', Figure 1). (S'') compares FENEON_GEN (GPT-4o rewrites in varied styles) with human FENEON_REF. If GPT-4o outputs have systematically different embedding dispersion from human texts for reasons unrelated to style, both inequalities can hold without any style effect, so the Section 5 conclusion that writing style influences embedding dispersion is not identifiable from these comparisons. A control that holds authorship and generation fixed, for example GPT-4o rewrites of the FENEON_REF texts in one uniform style versus GPT-4o rewrites in many styles, is needed.
  2. [§4.2, hypotheses (T') and (T'')] The topic hypotheses are also tested on human-versus-machine comparisons: (T') compares GPT-4o-generated FENEON_GEN to human QUENEAU_REF, and (T'') compares human FENEON_REF to GPT-4o-generated QUENEAU_GEN. The same authorship confound therefore applies, and the English QUENEAU_GEN texts are short by instruction. The only fully human comparison, (T-S), supports the ordering FENEON_REF > QUENEAU_REF, but it does not by itself disentangle topic from author and text-length differences. The paper would be substantially strengthened by a generated-control pair in which only topic variability changes while style and generation procedure are held fixed.
  3. [§4.3, Eqs. (8)-(9)] The interpretability analysis computes Pearson correlations between Δd(X,Y) = d_X(i) − d_Y(j) and Δf_s(X,Y) = f_X^s(i) − f_Y^s(j) without specifying a correspondence between the i-th vector of class X and the j-th vector of class Y. Since the two classes have no natural pairing, it is unclear how a single correlation is obtained; if all pairwise differences are pooled, the effective sample is larger than the 73 texts per class and the reported significance levels are not valid pairwise tests. This makes the feature-dispersion correlations in Figure 4 difficult to interpret and weakens the interpretability evidence for the style effect.
minor comments (6)
  1. [Figure 1] The French prompt for QUENEAU_GEN contains the typo 'Ré écris'; it should be 'Réécris'.
  2. [Section 3] The sentence 'we created a generated corpus by tranforming these original texts' contains a typo: 'tranforming' should be 'transforming'.
  3. [Section 4.1, Table 2] The reported ranking order '2D PCA, 3D PCA, 10D PCA, 5D PCA' is inconsistent with the mean scores in Table 2: for French, 10D PCA (0.6748) is higher than 3D PCA (0.6117) and even 2D PCA (0.6623); the text should describe the order actually implied by the combined means.
  4. [Table 2 and surrounding text] The model name is written 'xml-roberta-large' once in the paragraph after Table 2, while the table, the list of models, and the references use 'xlm-roberta-large'.
  5. [Section 4.2, Table 3] Significance levels are reported without any multiple-comparison correction across twelve models, two languages, and four local hypotheses; a note on the false-discovery rate or a justification for not correcting would improve the reliability of the pattern reported.
  6. [Limitations] The Limitations section does not mention the human-versus-machine confound that affects the style and topic hypotheses, nor the length constraint in the English QUENEAU_GEN prompt; both are central to interpreting the results and should be acknowledged.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the dispersion hypotheses are empirical comparisons, not fitted inputs; the only self-citations are non-load-bearing background references.

full rationale

The paper's derivation chain is not circular. Section 4.2 defines dispersion via Eq. (4) as a parameter-free mean Euclidean distance from the class centroid, and hypotheses (T), (S), and (T-S) are pre-specified directional inequalities among these measured quantities, e.g., dbar(QUENEAU_REF) > dbar(QUENEAU_GEN). No parameter is fitted to the dispersion data and then renamed a prediction; UMAP and PCA configurations are selected before the targeted comparisons, and the metric itself involves no fitted coefficients. The interpretability analysis uses the external Terreau et al. (2021) framework rather than fitting dispersion to stylistic features, so the style-feature correlations are not constructed from the dispersion outcome. The only self-citations (Faye et al., 2024; Icard et al., 2024, Section 2) are background references on lexical and punctuation style markers and are not load-bearing for the central claim. The reviewer-style concern that QUENEAU_GEN versus QUENEAU_REF, and FENEON_GEN versus FENEON_REF, differ in authorship and generation procedure as well as in style is a genuine threat to causal identifiability, but it is a confound, not a circular reduction by the paper's own equations. Therefore the central claim retains independent empirical content.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central claim rests on corpus construction and geometric assumptions rather than on a mathematical derivation. The free parameters listed are hand-chosen design values that affect the dispersion measurements; the axioms are the unverified conditions under which the comparisons isolate style and topic effects.

free parameters (2)
  • UMAP/PCA target dimension = 2 (dimensions)
    The paper selected 2D as the best dimension after computing mean S_D scores for clustering and after observing the strongest hypothesis validation for dispersion; this post-hoc choice contributes to the reported consistency.
  • Texts per class = 73
    Queneau's Exercices de style were truncated from 99 to 73 texts to balance the corpus with 73 Fénéon texts; this hand-chosen sample size may affect dispersion estimates.
assumptions (5)
  • domain assumption GPT-4o generation changes only the intended style/topic dimension while preserving the other dimension and all other textual properties
    Section 3.2 uses GPT-4o to rewrite Queneau in Fénéon style and Fénéon in Queneau styles, but no human evaluation or independent style classifier is provided; clustering validation uses the same embeddings.
  • domain assumption UMAP projections preserve enough global structure for Euclidean centroid distances to be meaningful measures of dispersion
    Section 4.2 computes d_iX in UMAP space; UMAP is nonlinear and does not preserve global distances exactly, which could distort the inter-class comparisons.
  • domain assumption The 73 texts within each class are independent observations for t-tests
    Generated texts are derived from the same source texts, introducing dependencies; t-tests and p-values in Tables 3 and 4 ignore this.
  • domain assumption English translations are stylistically equivalent to the French originals for the purpose of dispersion analysis
    English classes use translations by Wright and Sante; translation choices may alter function words, punctuation, and other features, as the paper itself notes in Section 4.3.
  • domain assumption The eight stylistic feature groups from Terreau et al. sufficiently capture style variation
    Section 4.3 relies on these features for interpretability; they may not cover all stylistic dimensions, as acknowledged in Limitations.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Embedding Style Beyond Topics: Analyzing Dispersion Effects Across Different Language Models." pith.science (2026). https://pith.science/paper/ZE7KS4U6

@misc{pith2026250100828,
  author       = {Pith},
  title        = {Pith review of: Embedding Style Beyond Topics: Analyzing Dispersion Effects Across Different Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZE7KS4U6}},
  note         = {Machine review of arXiv:2501.00828}
}
read the original abstract

This paper analyzes how writing style affects the dispersion of embedding vectors across multiple, state-of-the-art language models. While early transformer models primarily aligned with topic modeling, this study examines the role of writing style in shaping embedding spaces. Using a literary corpus that alternates between topics and styles, we compare the sensitivity of language models across French and English. By analyzing the particular impact of style on embedding dispersion, we aim to better understand how language models process stylistic information, contributing to their overall interpretability.

Figures

Figures reproduced from arXiv: 2501.00828 by the authors.

Figure 1
Figure 1. French and English GPT-4o prompts used for generating QUENEAU_GEN and FENEON_GEN, based on QUENEAU_REF and FENEON_REF. computational efficiency, explainability, and high performance according to the Massive Text Embedding Benchmark (MTEB) (Muennighoff et al., 2022) at time of the paper submission (September 16, 2024).2 The full list of tested models is presented in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. 2D PCA projection of the 4 clusters obtained with [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. 2D UMAP contour plots of the embedding dispersion obtained on the QUENEAU-FENEON corpus with model all-MiniLM-L12-v2, for French (left) and for English (right). In each subplot, the overall spread of the embeddings around centroid (for the last seed) is represented by the external contour line, the isolines represent differences in densities of embedding vectors, the centroid is indicated by a dot, and ¯dX correspon… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Correlation matrices between differences in [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 27 canonical work pages

  1. [1]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRING...

  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]

    Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774

    arXiv 2023

  4. [4]

    Aliakbar et al

    A. Aliakbar et al. 2022. Enhancing k-means clustering performance using pca. Journal of Data Science

    work page 2022

  5. [5]

    Farzana Anowar, Samira Sadaoui, and Bassant Selim. 2021. Conceptual and empirical comparison of dimensionality reduction algorithms (pca, kpca, lda, mds, svd, lle, isomap, le, ica, t-sne). Computer Science Review, 40:100378

    work page 2021

  6. [6]

    Janek Bevendorff, Xavier Bonet Casals, Berta Chulvi, Daryna Dementieva, Ashaf Elnagar, Dayne Freitag, Maik Fr \"o be, Damir Koren c i \'c , Maximilian Mayerl, Animesh Mukherjee, et al. 2024. Overview of pan 2024: multi-author writing style analysis, multilingual text detoxification, oppositional thinking analysis, and generative ai authorship verification...

    work page 2024

  7. [7]

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

    2003

  8. [8]

    Quang Anh Bui, Muriel Visani, Sophea Prum, and Jean-Marc Ogier. 2011. Writer identification using tf-idf for cursive handwritten word recognition. In 2011 International Conference on Document Analysis and Recognition, pages 844--848. IEEE

    work page 2011

Show all 45 references
  1. [9]

    Haoyang Chen, Zhongyuan Han, Zengyao Li, and Yong Han. 2023. A writing style embedding based on contrastive learning for multi-author writing style analysis. In CLEF 2023 Working Notes. CEUR Workshop Proceedings

    2023

  2. [10]

    Samuel Rhys Cox, Yunlong Wang, Ashraf Abdul, Christian Von Der Weth, and Brian Y. Lim. 2021. Directed diversity: Leveraging language embedding distances for collective creativity in crowd ideation. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems...

    2021

  3. [11]

    Ning Dai, Jianze Liang, Xipeng Qiu, and Xuanjing Huang. 2019. Style transformer: Unpaired text style transfer without disentangled latent representation. arXiv preprint arXiv:1905.05621

    2019 arXiv

  4. [12]

    Leon Danon, Albert Diaz-Guilera, Jordi Duch, and Alex Arenas. 2005. Comparing community structure identification. Journal of statistical mechanics: Theory and experiment, 2005(09):P09008

    2005

  5. [13]

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805

    2018 arXiv

  6. [14]

    Adji B Dieng, Francisco JR Ruiz, and David M Blei. 2020. Topic modeling in embedding spaces. Transactions of the Association for Computational Linguistics, 8:439--453

    2020

  7. [15]

    G \'e raud Faye, Benjamin Icard, Morgane Casanova, Julien Chanson, Fran c ois Maine, Fran c ois Bancilhon, Guillaume Gadek, Guillaume Gravier, and Paul \'E gr \'e . 2024. Exposing propaganda: an analysis of stylistic cues comparing human annotations and machine classification....

    2024

  8. [16]

    F \'e lix F \'e n \'e on. 2007. Novels in Three Lines. New York Review of Books. Translated by Luc Sante

    2007

  9. [17]

    F \'e lix F \'e n \'e on and Joan U Halperin. 1970. uvres plus que compl \`e tes , volume 2. Librairie Droz

    1970

  10. [18]

    Maarten Grootendorst. 2022. Bertopic: Neural topic modeling with a class-based tf-idf procedure. arXiv preprint arXiv:2203.05794

    2022 arXiv

  11. [19]

    Yaru Hao, Li Dong, Furu Wei, and Ke Xu. 2021. Self-attention attribution: Interpreting information interactions inside transformer. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 12963--12971

    2021

  12. [20]

    J Berenike Herrmann, Arthur M Jacobs, and Andrew Piper. 2021. Computational stylistics. Handbook of Empirical Literary Studies, pages 451--486

    2021

  13. [21]

    Holland et al

    P. Holland et al. 2020. Improving clustering interpretability with pca. Machine Learning Journal

    2020

  14. [22]

    Benjamin Icard, Fran c ois Maine, Morgane Casanova, G \'e raud Faye, Julien Chanson, Guillaume Gadek, Ghislain Atemezing, Fran c ois Bancilhon, and Paul \'E gr \'e . 2024. A multi-label dataset of french fake news: Human and machine insights. In Proceedings of the 2024 Joint I...

    2024

  15. [23]

    Albert Q Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, et al. 2023. Mistral 7b. arXiv preprint arXiv:2310.06825

    2023 arXiv

  16. [24]

    I. T. Jolliffe. 2016. Principal Component Analysis. Springer

    2016

  17. [25]

    JP Kincaid. 1975. Derivation of new readability formulas (automated readability index, fog count, and flesch reading ease formula) for navy enlisted personnel

    1975

  18. [26]

    Chang Liu, Zhongyuan Han, Haoyang Chen, and Qingbiao Hu. 2024. Team liuc0757 at pan: A writing style embedding method based on contrastive learning for multi-author writing style analysis. Working Notes of CLEF

    2024

  19. [27]

    Gonz \'a lez, and Thamar Solorio

    Suraj Maharjan, Deepthi Mave, Prasha Shrestha, Manuel Montes, Fabio A. Gonz \'a lez, and Thamar Solorio. 2019. Jointly learning author and annotated character n-gram embeddings: A case study in literary text. In Proceedings of the International Conference on Recent Advances in...

    2019

  20. [28]

    Inderjeet Mani. 2022. Computational modeling of narrative. Springer Nature

    2022

  21. [29]

    Christopher D Manning. 2008. Introduction to information retrieval

    2008

  22. [30]

    McInnes, J

    L. McInnes, J. Healy, and J. Melville. 2018. Umap: Uniform manifold approximation and projection for dimension reduction. arXiv preprint arXiv:1802.03426

    2018 arXiv

  23. [31]

    Niklas Muennighoff, Nouamane Tazi, Lo \" c Magne, and Nils Reimers. 2022. Mteb: Massive text embedding benchmark. arXiv preprint arXiv:2210.07316

    2022 arXiv

  24. [32]

    Nicole Peinelt, Dong Nguyen, and Maria Liakata. 2020. tbert: Topic models and bert joining forces for semantic similarity detection. In Proceedings of the 58th annual meeting of the association for computational linguistics, pages 7047--7055

    2020

  25. [33]

    Raymond Queneau. 1947. Exercices de style: Edition gallimard. Collection Folio

    1947

  26. [34]

    Raymond Queneau. 2013. Exercises in Style. New Directions Publishing. Translated by Barbara Wright

    2013

  27. [35]

    Nils Reimers and Iryna Gurevych. 2019. Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv preprint arXiv:1908.10084

    2019 arXiv

  28. [36]

    Germ \'a n R \' os-Toledo, Juan Pablo Francisco Posadas-Dur \'a n, Grigori Sidorov, and No \'e Alejandro Castro-S \'a nchez. 2022. Detection of changes in literary writing style using n-grams as style markers and supervised machine learning. Plos one, 17(7):e0267590

    2022

  29. [37]

    Claude Elwood Shannon. 1948. A mathematical theory of communication. The Bell system technical journal, 27(3):379--423

    1948

  30. [38]

    J. Shlens. 2014. A tutorial on principal component analysis. arXiv preprint arXiv:1404.1100

    2014 arXiv

  31. [39]

    Urvashi Soni and Sunita Dwivedi. 2024. Clutching of clustering validation criteria. International Journal of Future Computer and Communication, 13(1)

    2024

  32. [40]

    Enzo Terreau, Antoine Gourru, and Julien Velcin. 2021. Writing style author embedding evaluation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 84--93. Association for Computational Linguistics

    2021

  33. [41]

    Gaurav Verma and Balaji Vasan Srinivasan. 2019. A lexical, syntactic, and semantic perspective for understanding style in text. arXiv preprint arXiv:1909.08349

    2019 arXiv

  34. [42]

    Anna Wegmann, Marijn Schraagen, and Dong Nguyen. 2022. Same author or just same topic? towards content-independent style representations. In Proceedings of the 7th Workshop on Representation Learning for NLP, pages 249--268. Association for Computational Linguistics

    2022

  35. [43]

    H. Wold. 1987. Principal component analysis. Technometrics, 38(3):235--238

    1987

  36. [44]

    C Udny Yule. 2014. The statistical study of literary vocabulary. Cambridge University Press

    2014

  37. [45]

    Haiyan Zhao, Hanjie Chen, Fan Yang, Ninghao Liu, Huiqi Deng, Hengyi Cai, Shuaiqiang Wang, Dawei Yin, and Mengnan Du. 2024. Explainability for large language models: A survey. ACM Transactions on Intelligent Systems and Technology, 15(2):1--38

    2024

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.