Pith. sign in

REVIEW 4 major objections 5 minor 32 references

Isolating authorship from content with semantic embeddings and contrastive learning

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Frozen content negatives sharpen authorship attribution by pushing style embeddings away from topic.

desk verdict The core idea is plausible and honestly reported, but the main comparison is confounded by the baseline using fewer InfoNCE negatives, so the headline gains may not come from disentanglement. read the letter →

arxiv 2411.18472 v1 pith:2AMBXDSW submitted 2024-11-27 cs.CL cs.LG

Javier Huertas-Tato , Adrián Girón-Jiménez , Alejandro Martín , David Camacho This is my paper · ORCID
classification cs.CLcs.LG
keywords authorshipattributionstyle-contentdisentanglementcontrastivelearningInfoNCEsemanticembeddingshardnegativeszero-shottransfertransformerfine-tuning
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

This paper argues that authorship embeddings trained with contrastive learning are contaminated by topic: because authors tend to write about the same subjects, the model learns spurious content correlations. The authors propose a modified InfoNCE loss that, alongside the usual same-author positives and different-author negatives, adds negatives produced by a frozen semantic-similarity model, pushing the trainable style space away from a fixed content space. On two datasets the procedure improves attribution accuracy, with the largest gains (up to 10 percentage points) in the hard case where prolific authors write across many topics, and it reduces the share of errors caused by confusion within one topic. The method also preserves more zero-shot ability on PAN challenges than plain fine-tuning does.

What carries the argument

The central object is the modified InfoNCE objective of Eq. (2) and Algorithm 1. For two document sets $D_A$ and $D_B$, the reference set is $R(D_A, D_B) = s(D_B) \parallel c(D_A) \parallel c(D_B)$, so each row of the cosine-similarity matrix contains one same-author positive and $3N-1$ negatives, including the content embeddings of the anchor text itself. The frozen content model (UAE-Large-V1) fixes the content subspace, and the trainable style model (warm-started from STAR) is pushed away from it by the cross-entropy loss. This geometric separation is the entire mechanism: the only learnable parameters are in the style encoder.

What would settle it

Train the same loss with a content model that is deliberately style-aware (for example, a model fine-tuned on authorship) and check whether accuracy gains shrink or reverse; or measure how much authorship signal a classifier can recover from the frozen content embeddings alone — if that signal is high, the repulsion is pushing against a space that already contains style.

Watch

Extended reading notes

Core claim

The paper's central claim is that style-content entanglement in authorship representations can be reduced by extending InfoNCE with hard negatives drawn from a frozen semantic embedding space. Concretely, for each document the loss treats same-author embeddings as positives and, as negatives, different-author style embeddings, content embeddings of the same text, and content embeddings of other texts; minimizing the cross-entropy then separates the style subspace from the content subspace. The paper reports that this 'detangled' fine-tuning beats contrastive fine-tuning under identical conditions and training time, and that on the most heterogeneous-author subsets the accuracy gain reaches about 10 percentage points while topic-related miss rates fall. It also reports better retention of zero-shot attribution performance than plain fine-tuning, especially when the disentangling dataset is large.

Load-bearing premise

The method assumes that the frozen semantic model's embedding space contains essentially no stylistic information, so pushing style embeddings away from it removes content rather than distorting style; the paper itself admits this is not strictly true.

Editorial extensions

If this is right

  • On the blog dataset, detangled fine-tuning raises attribution accuracy from 0.3928 (simple fine-tune) to 0.4469 and reduces the same-topic miss proportion from 0.1193 to 0.1031.
  • On the fan-fiction dataset, detangled fine-tuning reaches 0.7924 accuracy versus 0.7741 for simple fine-tuning, and on the most topic-diverse author subsets it outperforms zero-shot STAR by roughly 8 to 11 percentage points.
  • Across PAN11-19 challenges, the detangled fan-fiction model achieves the highest average accuracy (0.6346) among the fine-tuned variants, while plain fan-fiction fine-tuning drops to 0.5748, suggesting the method preserves zero-shot ability better than ordinary fine-tuning.
  • Because the method is a regularizer rather than a new architecture, it applies to any contrastively trained style encoder paired with any frozen semantic encoder of matching dimension.
  • The detangled model consistently lowers topic-related attribution failures, which is direct evidence that the learned style space depends less on content.

Reading between the lines

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

  • Beyond the paper, the same loss could be applied to any contrastively trained style encoder, so the disentanglement recipe is not specific to authorship attribution; tasks like author profiling or style-transfer evaluation could use the identical frozen-content negative trick.
  • Beyond the paper, swapping the frozen content model — for example to a different semantic-similarity encoder or to an MLM embedding — would test how much of the gain depends on the specific blind spots of UAE rather than on content removal generally.
  • Beyond the paper, because the PAN19 challenge overlaps the fanfiction training corpus, the strong zero-shot result on PAN19 may partly reflect domain overlap; a held-out-domain evaluation with no overlap would give a cleaner estimate of preserved zero-shot ability.
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 / 5 minor

Summary. The paper proposes a modification of the InfoNCE contrastive objective for authorship attribution: in addition to the usual in-batch style negatives, the loss adds frozen semantic-embedding negatives produced by a content model (UAE-Large-V1), so that the style embedding space is pushed away from the content embedding space. The method is evaluated on a blog dataset, a large fanfiction corpus, and the PAN 11-19 authorship challenges, comparing against RoBERTa, a zero-shot style model (STAR), and a simple InfoNCE fine-tuning of STAR. The paper reports consistent accuracy gains for the proposed 'detangled' model, with the largest improvements (up to about 10 points) on a subset of highly topic-diverse authors, and a reduction in the fraction of attribution errors that confuse authors writing about the same topic.

Significance. If the central claim were fully supported, the proposal would be a simple and inexpensive regularizer that reduces content leakage in authorship embeddings, with practical value for authorship attribution in topic-diverse settings. The paper evaluates on multiple datasets and standard challenges, which is a strength. However, the current empirical support is weakened by a confounded baseline: the 'detangled' loss uses three times as many negatives as the simple InfoNCE baseline, and the paper's own disentanglement metric (same-topic-when-miss) does not isolate the proposed mechanism because it improves for the simple fine-tuning baseline as well. The assumption that the frozen content model is free of style is acknowledged as false but is not tested through an alternative content-model ablation.

major comments (4)
  1. [Section 5.4, Eq. (2), Algorithm 1] The comparison between 'Fine-tune (simple)' and 'Fine-tune (detangle)' is not controlled for the number of negative examples. In Algorithm 1 and Eq. (2), the reference set R(DA, DB) = s(DB) || c(DA) || c(DB) gives a denominator of 3N terms per anchor, whereas the simple InfoNCE baseline has only the N style embeddings in the batch as negatives. InfoNCE is known to improve as the negative set grows, so the accuracy gains in Tables 2, 4, and 5 could be due to a larger negative pool or stronger contrastive regularization rather than to semantically informed disentanglement. The claim in Section 5.4 that the detangled model 'outperforms a contrastive fine-tuning under the same conditions and training time' requires a controlled comparison, for example a simple InfoNCE baseline with additional style negatives (or sampled negatives) to match the 3N pool size, or a detangled variant that subsamples the combined reference set to N negatives.
  2. [Section 3, Eq. (3)] The loss definition in Eq. (3) is mathematically incorrect as written: LCE sums log P over all 3N reference items, but the standard InfoNCE / cross-entropy loss for a single positive is -log P(positive). This is inconsistent with Algorithm 1, which calls cross_entropy(simf, labels) and thus correctly implements the standard loss. The equation needs to be corrected, otherwise the method's derivation in Section 3 does not match the implemented objective.
  3. [Tables 2, 4, and 5] The reported results do not include standard deviations, confidence intervals, or significance tests. The paper states that evaluation is repeated 10 times for stability, but this appears to be 10 random selections of validation documents rather than 10 independent training runs. With only one training run per condition, the observed improvements of 2-4 points (and even the larger gains in Table 5) could be within run-to-run variance. The authors should report variance across multiple random seeds, or at least provide bootstrap confidence intervals for the evaluation metric.
  4. [Table 3, Section 5.1] The 'Same topic when miss' metric decreases monotonically from RoBERTa (0.1382) to STAR (0.1258) to Fine-tune simple (0.1193) to Fine-tune detangle (0.1031). This monotonic trend means the metric does not isolate the proposed disentanglement mechanism: simple fine-tuning already reduces topic-related errors, and the detangled model's lower value could simply be a continuation of that trend. To support the claim that the semantic-embedding negatives specifically reduce topic-driven confusions, a comparison that matches training conditions and controls for the number of negatives is needed.
minor comments (5)
  1. [Algorithm 1] In Algorithm 1, lines for sim2 and sim3 use 'DA' and 'DB' as inputs to cosine_distance, but they should be the content embeddings CA and CB; as written, these lines are dimensionally inconsistent with the text of Section 3.
  2. [Eq. (2)] The term Scos(DA,i) in Eq. (2) should be Scos(SA,i), where SA,i = s(DA,i), since the similarity is computed between style embeddings.
  3. [Section 4.3] The sentence 'Our code is available at lightning.ai/docs/pytorch/stable/' points to the PyTorch Lightning documentation rather than to a code repository for this paper; the authors should provide an actual repository link or remove this claim.
  4. [Table 4 caption] The caption of Table 4 says 'Attribution results across the entire blog author set', but the table presents fanfiction results; the caption should be corrected.
  5. [Abstract and throughout] The paper consistently uses 'detangle' and 'detangling' where 'disentangle' and 'disentangling' are standard; this is a wording issue, not a technical one, but it should be normalized for clarity.

Circularity Check

1 steps flagged · score 2.0 of 10

Minor self-definitional element: style-content repulsion is the training objective itself, but the core attribution-accuracy claim rests on held-out evaluation against independent baselines.

  1. self definitional [Section 3 (Eq. 2) and Algorithm 1; cited again in Section 5.4]
    "To counteract this, we modify our objective function to differentiate style from content entirely. We maximize similarity when texts belong to the same author, while minimizing similarity from texts belonging to different authors, while simultaneously maximizing distance between embeddings generated by the style model from embeddings generated by the content model. ... This twofold demonstration offers strong evidence that the model obtains higher quality stylistic features by distancing itself from the semantic embedding space."

    In Algorithm 1, simf = cat(sim1, sim2, sim3) with labels = range(0,N); for each style anchor row, the only positive is its paired style embedding, so every content embedding c(DA,j) and c(DB,j) is included as a negative in the cross-entropy denominator of Eq. 2. Any decrease in style-content similarity is therefore a direct consequence of the loss, not an empirical finding. The Sec. 5.4 statement that the model improves 'by distancing itself from the semantic embedding space' thus restates the objective. This is only a minor circularity because the headline attribution-accuracy numbers come from held-out author evaluation against RoBERTa, STAR, and simple InfoNCE baselines, and do not reduce to Eq. 2 alone.

full rationale

The central claim, that the detangled fine-tune improves authorship attribution accuracy and reduces same-topic errors, is tested on held-out authors, with the blog and fan-fiction results compared against RoBERTa, zero-shot STAR, and a standard InfoNCE fine-tune, plus PAN11-19 challenges. Those results cannot be derived from the loss definition; they depend on learned representations. The raw repulsion between style and content spaces is, however, written directly into the loss (Eq. 2 and Algorithm 1), so presenting 'distancing from the semantic embedding space' as evidence is self-definitional, though not load-bearing for the main accuracy claim. Self-citations to PART and STAR supply the base model and baselines but carry no uniqueness or correctness argument, so they are not circular. The acknowledged imperfection of UAE (Section 3) and the uncontrolled negative-pool size in the 'same conditions' comparison (3N vs N negatives) are genuine experimental limitations, but they are confounds, not circularity.

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

The ledger is dominated by domain assumptions about embedding geometry rather than invented entities. The method depends on the unverified premise that a frozen semantic model defines a style-free content axis, and on the choice of several training hyperparameters. No new physical or conceptual entities are introduced.

free parameters (4)
  • Trainable temperature tau
    Scaling temperature in Eq. 2 is trained during fine-tuning; it controls the sharpness of the InfoNCE softmax and is part of the model, not a reported constant.
  • Learning rate per dataset = 0.001 (fanfiction), 0.005 (blog)
    Section 4.3 states learning rate was 'adjusted for optimal performance from a set of {0.001, 0.002, 0.005, 0.01}', making the comparison partly tuned to each dataset.
  • Epochs per dataset = 25 (fanfiction), 250 (blog)
    Section 4.3 states 'epochs have been adjusted for dataset size', so training length is a hand-chosen free parameter.
  • Token chunk length = 512
    Section 4.2 uses randomized truncation to 512 tokens for training; this choice affects what the style and content models see and is not derived from any principle.
assumptions (4)
  • domain assumption Authorship/style can be represented as a subspace of embedding space and cosine distance captures style similarity.
    Section 3 and Figure 1 assume a style embedding space can be separated from a content embedding space using cosine distance in the shared output space of STAR and UAE.
  • domain assumption The frozen UAE-Large-V1 semantic model produces content embeddings that are sufficiently free of style to act as a valid content reference.
    Section 3 states the disentanglement 'is strictly reliant on the content embedding model's capabilities' and admits the assumption is 'obviously untrue'.
  • domain assumption Pushing style embeddings away from content embeddings in InfoNCE yields better style embeddings rather than merely distorting the space.
    Section 3.1 defines the loss with content negatives; the paper provides no formal argument that repulsion from a fixed reference transfers to unseen authors.
  • standard math Standard machinery of contrastive learning: InfoNCE, softmax cross-entropy, and cosine similarity behave as expected on high-dimensional transformer embeddings.
    Eqs. 1-4 and Algorithm 1 rely on the standard InfoNCE formulation from Oord et al. [2], treated as background.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Isolating authorship from content with semantic embeddings and contrastive learning." pith.science (2026). https://pith.science/paper/2AMBXDSW

@misc{pith2026241118472,
  author       = {Pith},
  title        = {Pith review of: Isolating authorship from content with semantic embeddings and contrastive learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2AMBXDSW}},
  note         = {Machine review of arXiv:2411.18472}
}
read the original abstract

Authorship has entangled style and content inside. Authors frequently write about the same topics in the same style, so when different authors write about the exact same topic the easiest way out to distinguish them is by understanding the nuances of their style. Modern neural models for authorship can pick up these features using contrastive learning, however, some amount of content leakage is always present. Our aim is to reduce the inevitable impact and correlation between content and authorship. We present a technique to use contrastive learning (InfoNCE) with additional hard negatives synthetically created using a semantic similarity model. This disentanglement technique aims to distance the content embedding space from the style embedding space, leading to embeddings more informed by style. We demonstrate the performance with ablations on two different datasets and compare them on out-of-domain challenges. Improvements are clearly shown on challenging evaluations on prolific authors with up to a 10% increase in accuracy when the settings are particularly hard. Trials on challenges also demonstrate the preservation of zero-shot capabilities of this method as fine tuning.

Figures

Figures reproduced from arXiv: 2411.18472 by the authors.

Figure 1
Figure 1. Disentangling embedding spaces for style and content [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Contrastive learning objective for disentanglement [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Contrastive learning objective for disentanglement [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Accuracy on attribution when authors publish across the same category [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 26 canonical work pages

  1. [1]

    Authorship attribution in the era of llms: Problems, methodologies, and challenges

    Baixiang Huang, Canyu Chen, and Kai Shu. Authorship attribution in the era of llms: Problems, methodologies, and challenges. arXiv preprint arXiv:2408.08946, 2024

    arXiv 2024

  2. [2]

    Representation learning with contrastive predictive coding

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

    arXiv 2018

  3. [3]

    A survey of machine learning-based author profiling from texts analysis in social networks

    Sarra Ouni, Fethi Fkih, and Mohamed Nazih Omri. A survey of machine learning-based author profiling from texts analysis in social networks. Multimedia Tools and Applications, 82(24):36653–36686, 2023

    work page 2023

  4. [4]

    Jacob Tyo, Bhuwan Dhingra, and Zachary C. Lipton. On the State of the Art in Authorship Attribution and Authorship Verification, October 2022

    work page 2022

  5. [5]

    A random forest approach for authorship profiling

    Alonso Palomino-Garibay, Adolfo T Camacho-Gonzalez, Ricardo A Fierro-Villaneda, Irazu Hernandez-Farias, Davide Buscaldi, Ivan V Meza-Ruiz, et al. A random forest approach for authorship profiling. In Proceedings of CLEF, 2015

    work page 2015

  6. [6]

    Gender identification in twitter using n-grams and lsa

    Saman Daneshvar and Diana Inkpen. Gender identification in twitter using n-grams and lsa. In proceedings of the ninth international conference of the CLEF association (CLEF 2018) . CEUR-WS, 2018

    work page 2018

  7. [7]

    Writeprints: A stylometric approach to identity-level identification and similarity detection in cyberspace

    Ahmed Abbasi and Hsinchun Chen. Writeprints: A stylometric approach to identity-level identification and similarity detection in cyberspace. ACM Trans. Inf. Syst., 26(2):7:1–7:29, April 2008

    work page 2008

  8. [8]

    Cross-genre author profile prediction using stylometry-based approach

    Shaina Ashraf, Hafiz Rizwan Iqbal, and Rao Muhammad Adeel Nawab. Cross-genre author profile prediction using stylometry-based approach. In CLEF (Working Notes), pages 992–999, 2016

    work page 2016

Show all 32 references
  1. [9]

    A four feature types approach for detecting bot and gender of twitter users

    Johan Fernquist. A four feature types approach for detecting bot and gender of twitter users. In CLEF (Working Notes), 2019

    2019

  2. [10]

    In Claude Sammut and Geoffrey I

    TF–IDF. In Claude Sammut and Geoffrey I. Webb, editors, Encyclopedia of Machine Learning , pages 986–987. Springer US, Boston, MA, 2010

    2010

  3. [11]

    Continuous N-gram Representations for Authorship Attribu- tion

    Yunita Sari, Andreas Vlachos, and Mark Stevenson. Continuous N-gram Representations for Authorship Attribu- tion. In Mirella Lapata, Phil Blunsom, and Alexander Koller, editors, Proceedings of the 15th Conference of the European Chapter of the Association for Computational Lin...

    2017

  4. [12]

    Authorship Attribution Using Word Sequences

    Rosa María Coyotl-Morales, Luis Villaseñor-Pineda, Manuel Montes-y Gómez, and Paolo Rosso. Authorship Attribution Using Word Sequences. In José Francisco Martínez-Trinidad, Jesús Ariel Carrasco Ochoa, and Josef Kittler, editors, Progress in Pattern Recognition, Image Analysis ...

    2006

  5. [13]

    Authorship verification machine learning methods for style change detection in texts

    Gianni X Jacobo, Valeria Dehesa-Corona, Ariel D Rojas-Reyes, and Helena Gómez-Adorno. Authorship verification machine learning methods for style change detection in texts. In CLEF (Working Notes) , pages 2652–2658, 2023

    2023

  6. [14]

    Authorship Attribution in Fan-Fictional Texts given variable length Character and Word N-Grams

    Lukas Muttenthaler, Gordon Lucas, and Janek Amann. Authorship Attribution in Fan-Fictional Texts given variable length Character and Word N-Grams. In CLEF (Working Notes), 2019. 13 Isolating authorship from content with semantic embeddings and contrastive learning

    2019

  7. [15]

    Contrastive disentanglement for authorship attribution

    Zhiqiang Hu, Thao Thanh Nguyen, Yujia Hu, Chia-Yu Hung, Ming Shan Hee, Chun Wei Seah, and Roy Ka- Wei Lee. Contrastive disentanglement for authorship attribution. In Companion Proceedings of the ACM Web Conference 2024, pages 1657–1666, Singapore Singapore, May 2024. ACM

    2024

  8. [16]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems , 30, 2017

    2017

  9. [17]

    Bert: Pre-training of deep bidirectional transformers for language understanding

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langu...

    2019

  10. [18]

    Bertaa: Bert fine-tuning for authorship attribution

    Maël Fabien, Esaú Villatoro-Tello, Petr Motlicek, and Shantipriya Parida. Bertaa: Bert fine-tuning for authorship attribution. In Proceedings of the 17th International Conference on Natural Language Processing (ICON) , pages 127–137, 2020

    2020

  11. [19]

    Part: Pre-trained authorship representation transformer

    Javier Huertas-Tato, Alvaro Huertas-Garcia, Alejandro Martin, and David Camacho. Part: Pre-trained authorship representation transformer. Human-centric Computing and Information Sciences , 14, 2024

    2024

  12. [20]

    Roberta: A robustly optimized bert pretraining approach

    Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692, 2019

    1907 arXiv

  13. [21]

    Adversarial decomposition of text representation, April 2019

    Alexey Romanov, Anna Rumshisky, Anna Rogers, and David Donahue. Adversarial decomposition of text representation, April 2019

    2019

  14. [22]

    Generative Adversarial Nets

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative Adversarial Nets. In Advances in Neural Information Processing Systems , volume 27. Curran Associates, Inc., 2014

    2014

  15. [23]

    Same author or just same topic? towards content- independent style representations

    Anna Wegmann, Marijn Schraagen, and Dong Nguyen. Same author or just same topic? towards content- independent style representations. arXiv preprint arXiv:2204.04907, 2022

    2022 arXiv

  16. [24]

    Topic-regularized authorship representation learning

    Jitkapat Sawatphol, Nonthakit Chaiwong, Can Udomcharoenchaikit, and Sarana Nutanong. Topic-regularized authorship representation learning. In Yoav Goldberg, Zornitsa Kozareva, and Yue Zhang, editors,Proceedings of the 2022 Conference on Empirical Methods in Natural Language Pr...

    2022

  17. [25]

    Understanding writing style in social media with a supervised contrastively pre-trained transformer

    Javier Huertas-Tato, Alejandro Martín, and David Camacho. Understanding writing style in social media with a supervised contrastively pre-trained transformer. Knowledge-Based Systems, 296:111867, 2024

    2024

  18. [26]

    Angle-optimized text embeddings

    Xianming Li and Jing Li. Angle-optimized text embeddings. arXiv preprint arXiv:2309.12871, 2023

    2023 arXiv

  19. [27]

    Effects of Age and Gender on Blogging

    Moshe Koppel Jonathan Schler. Effects of Age and Gender on Blogging. AAAI, October 2023

    2023

  20. [28]

    [Online; accessed 16

    Fanfiction.net Re-pack : Free Download, Borrow, and Streaming : Internet Archive, March 2014. [Online; accessed 16. Sep. 2024]

    2014

  21. [29]

    Overview of the author identification task at pan 2013

    Patrick Juola and Efstathios Stamatatos. Overview of the author identification task at pan 2013. CLEF (Working Notes), 1179, 2013

    2013

  22. [30]

    Overview of the author identification task at pan 2014

    Efstathios Stamatatos, Walter Daelemans, Ben Verhoeven, Martin Potthast, Benno Stein, Patrick Juola, Miguel A Sanchez-Perez, Alberto Barrón-Cedeño, et al. Overview of the author identification task at pan 2014. In CEUR Workshop Proceedings, volume 1180, pages 877–897. CEUR-WS, 2014

    2014

  23. [31]

    Overview of the author identification task at pan-2018: cross-domain authorship attribution and style change detection

    Mike Kestemont, Michael Tschuggnall, Efstathios Stamatatos, Walter Daelemans, Günther Specht, Benno Stein, and Martin Potthast. Overview of the author identification task at pan-2018: cross-domain authorship attribution and style change detection. In Working Notes Papers of th...

    2018

  24. [32]

    Overview of the cross-domain authorship attribution task at {PAN} 2019

    Mike Kestemont, Efstathios Stamatatos, Enrique Manjavacas, Walter Daelemans, Martin Potthast, and Benno Stein. Overview of the cross-domain authorship attribution task at {PAN} 2019. In Working Notes of CLEF 2019-Conference and Labs of the Evaluation F orum, Lugano, Switzerlan...

    2019

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.