What's Taboo for You? - An Empirical Evaluation of LLMs Behavior Toward Sensitive Content

Alfio Ferrara; Alice Tuveri; Elisabetta Rocchetti; Laura Pinnavaia; Sergio Picascia; Vojimir Ranitovic

REVIEW 5 major objections 5 minor 38 references

What's Taboo for You? - An Empirical Evaluation of LLMs Behavior Toward Sensitive Content

T0 review · 5 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read GPT-4o-mini systematically flattens sensitive language when asked only to paraphrase it.

desk verdict A solid empirical measurement of a real effect, but the annotation layer is underdocumented and the magnitude should be read with caution until those details are released. read the letter →

arxiv 2507.23319 v1 pith:5KAAAYIG submitted 2025-07-31 cs.CL

Alfio Ferrara , Sergio Picascia , Laura Pinnavaia , Vojimir Ranitovic , Elisabetta Rocchetti , Alice Tuveri This is my paper · ORCID

classification cs.CL

keywords implicitcontentmoderationLLMparaphrasingsanitizationsensitivelanguagesensitivityannotationGPT-4o-minizero-shotclassificationtextdetoxification

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

GPT-4o-mini, when asked only to paraphrase a sentence, tends to return a less sensitive version of it, even though no instruction to soften or detoxify was given. The paper claims this implicit moderation is systematic: expert annotators classified original and paraphrased sentences into four sensitivity levels, and the shifts consistently point downward, from Taboo and Derogatory toward Informal and Formal/Polite. The strongest evidence is the confusion matrix in which 51.6% of originally Taboo sentences become Informal after paraphrase, and Bowker's symmetry test rejects the null hypothesis ($\chi^2 = 138.255$, df = 6, p < 0.0001). The paper also reports that zero-shot open-source LLMs are worse than simple trained classifiers at reproducing expert sensitivity judgments, so automated moderation monitoring is not yet safe to hand to generic LLMs. If correct, the finding matters because aligned models may sanitize language by default in tasks where users expect neutral rewording, silently changing the content being processed.

What carries the argument

The machinery is a four-class ordinal sensitivity schema (Formal/Polite, Informal, Derogatory, Taboo) applied identically to original sentences and to GPT-4o-mini's paraphrases. Shifts are summarized in a 4×4 confusion matrix whose asymmetry is tested by Bowker's test for symmetry, and quantified by $\Delta$, the average signed change in sensitivity level (1 to 4) from original to paraphrase for each starting class. The schema does the work of making an otherwise fuzzy judgment — how offensive is this sentence? — measurable, so implicit moderation becomes a detectable, statistically testable pattern rather than an impression.

What would settle it

Recompute the confusion matrix and Bowker's test using only sentences where the two judges independently gave the same label; if the asymmetry disappears or the average drop for taboo sentences vanishes, the result is an artifact of how labels were combined rather than a real behavior of the model.

Watch

Extended reading notes

Core claim

The paper's central claim is that GPT-4o-mini performs implicit content moderation during paraphrase generation: it systematically lowers the sensitivity level of the text it rewrites. Using a four-level schema (Formal/Polite, Informal, Derogatory, Taboo) applied by ten expert annotators to 2,269 original and paraphrased sentences, the authors find that sensitivity is preserved for less sensitive categories (87.1% of Formal/Polite stay Formal/Polite) but collapses for the most sensitive ones: only 15.6% of Taboo sentences remain Taboo, 51.6% become Informal, and 58.9% of Derogatory sentences become Formal/Polite or Informal. The average shift metric $\Delta$ is $-1.7$ for original Taboo sentences and $-0.87$ for Derogatory, meaning roughly one to two sensitivity levels of reduction, and Bowker's test ($\chi^2 = 138.255$, df = 6, p < 0.0001) rejects symmetry of the shift matrix. A secondary claim is that zero-shot open-source LLMs do not reliably replicate expert sensitivity labels: traditional classifiers such as MLP and the Born classifier reach higher F1 scores, and no automated method exceeds roughly 0.53 overall F1.

Load-bearing premise

The load-bearing premise is that the human judges' labels are correct; the paper does not say how often the two judges disagreed, how disagreements were settled into a final label, or why 231 of the sampled sentences ended up without annotations. If those labels are unreliable, every measured shift inherits that unreliability.

Editorial extensions

If this is right

A neutral 'paraphrase this sentence' request to GPT-4o-mini is not content-preserving in sensitivity: taboo and derogatory material is routinely downgraded, so downstream uses of such paraphrases inherit a sanitization bias.
The direction of the shift is consistently toward less sensitive classes for every classifier that detects a shift, reinforcing the conclusion that aligned LLMs moderate by default rather than by instruction.
Because even the best automated classifier reaches only 0.53 overall F1, zero-shot or lightly trained models are not yet reliable substitutes for expert judgment in moderation monitoring.
Traditional classifiers trained on expert labels approximate expert sensitivity shifts more closely than zero-shot open-source LLMs, offering a cheaper local alternative for monitoring paraphrase behavior at scale.

Reading between the lines

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

If the same protocol were run on all three paraphrases instead of one per original sentence, the $\Delta$ estimates could be tested for stability across paraphrase draws; large variance would mean the one-sample annotation design understates the range of possible sanitization.
Treating the four sensitivity levels as evenly spaced is a modeling choice; a weighting that makes the Taboo-to-Derogatory distance larger than other gaps would change the magnitude of $\Delta$ but not the direction of the shift.
A natural next experiment is to prompt GPT-4o-mini with an explicit instruction to preserve tone and register; if the negative $\Delta$ shrinks, that would confirm the sanitization is an alignment default rather than an unavoidable property of paraphrasing.

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.

Desk Editor's Note

read the letter

Quick take: this paper gives a useful measurement of a real phenomenon—GPT-4o-mini systematically softens sensitive language during paraphrasing—but the measurement layer has a transparency gap. I would not bet against the effect, but I would not quote the headline magnitudes without seeing the annotation details.

What is new: the paper quantifies implicit sanitization in a paraphrase task using expert labels, and compares six open-source LLMs against six traditional classifiers. The main result is internally consistent: 51.6% of Taboo originals become Informal, 58.9% of Derogatory originals become Formal/Polite or Informal, and Bowker's test rejects symmetry with p < 0.0001. That is a genuine empirical result, not a demo. The classifier comparison is also useful: traditional methods, especially MLP and Born Classifier, track expert Delta better than the open LLMs, while overall accuracy is poor enough to caution against using generic LLMs for moderation automation.

Soft spots: Section 3.2 samples 2500 sentences but reports annotations for only 2269, with no explanation for the missing 231. More importantly, there is no inter-annotator agreement and no stated adjudication rule. With ten annotators and two independent labels per sentence, that is a standard and fixable transparency gap. If disagreements were resolved toward the less sensitive class for paraphrases, the effect could be inflated. I see no evidence that happened, but I cannot rule it out. There is also no control condition, so part of the shift could be ordinary paraphrase behavior rather than alignment-specific sanitization. The authors say the result 'suggests' implicit moderation, which is honest, but the design does not isolate the mechanism. No code or data release makes the measurement hard to verify. These are not fatal—the effect is large enough that even noisy labels would likely show it—but they cap the precision of the claim.

The citation pattern is fine. Born Classifier is a self-citation, but it is evaluated as a baseline against external human labels, not used to prove the main effect.

Recommendation: worth refereeing. A serious editor should send this out, with the request that the authors document the annotation process, add a control baseline, and release code/data. This is a paper for content moderation researchers, alignment evaluators, and anyone building benchmark tasks for sensitivity classification.

Referee Report

5 major / 5 minor

Summary. The paper investigates whether GPT-4o-mini implicitly moderates sensitive content when asked to perform a neutral paraphrase task. The authors collect 23,347 sentences containing sensitive expressions, paraphrase each three times, and then have 10 linguist experts annotate a subset of 1,250 original/paraphrase pairs into four sensitivity categories (Formal/Polite, Informal, Derogatory, Taboo). They report a confusion matrix showing that paraphrases tend to move from Taboo and Derogatory toward less sensitive categories, supported by Bowker's test of symmetry (χ²=138.255, df=6, p<0.0001), and compute a per-class average sensitivity shift Δ showing a mean drop of 1.7 levels for Taboo sentences. The paper also evaluates six open-source LLMs and six traditional classifiers on the same annotation task, finding that traditional classifiers such as MLP and Born Classifier match expert labels better than the zero-shot LLMs, with overall F1 around 0.5. The central conclusion is that GPT-4o-mini systematically sanitizes language even without explicit detoxification instructions.

Significance. If the central claim holds, the paper provides a concrete, quantitative demonstration that a proprietary aligned LLM performs implicit content moderation during a task that does not ask for it. This is a valuable contribution to the study of alignment side-effects and complements prior work that focuses on explicit detoxification or refusal behavior. The paper's strengths include a simple and reproducible experimental pipeline, the use of a neutral paraphrase prompt, a statistically significant test for asymmetric shifts, and the inclusion of diverse automated classifiers as a comparison. The dataset itself, even if limited in size, is a useful resource for studying sensitivity perception in LLMs. The main weakness is that the reliability of the expert-annotation ground truth is not documented, and several procedural details are missing, which currently prevents a full endorsement of the quantitative conclusions.

major comments (5)

[§3.2] The annotation layer is not sufficiently documented to support the load-bearing ground truth. The paper states that each sentence received two independent annotations, but it does not report inter-annotator agreement (e.g., Cohen's kappa or Krippendorff's alpha), does not describe how disagreements were resolved into a final label, and does not explain why only 2269 of the 2500 selected sentences have annotations. If annotations are noisy or if the aggregation rule correlates with whether a sentence is original or paraphrased, the confusion matrix in Figure 2, the Bowker statistic, and all Δ values in Table 1 inherit that bias. The 231 missing sentences also require an explanation, because their systematic exclusion (for instance, due to the 'Difficult to say' option) could shift the sensitivity distribution and distort the measured transition probabilities.
[§3.1] The handling of the 232 refusal responses from GPT-4o-mini is unspecified. The paper says these refusals 'were rare' but does not state whether a refused sentence was paraphrased again, whether the refusal was treated as a non-response, or whether such sentences were excluded from the annotation sample. Since the annotation section says 'we randomly selected one of its three paraphrases,' this presupposes that all original sentences have three valid paraphrases. If refusals are concentrated among the most offensive sentences, the annotated sample could be biased away from the top of the sensitivity scale, which would directly affect the estimated Δ for Taboo and Derogatory classes. The authors should report the exact handling of refusals and, if any sentences were excluded, state how this affects the representativeness of the sample.
[§3.3 and Table 1] It is unclear on which data split the Δ values for the traditional classifiers are computed. Section 3.2 describes an 80/20 train/test split for the traditional classifiers, but Table 1 reports Δ for each sensitivity class without specifying whether the predictions come from the 20% test set or the full dataset. If the full dataset is used, the classifiers have already seen 80% of the labels during training, making the Δ values and the ΔMSD comparisons optimistically biased. If only the test set is used, the per-class sample sizes are small and would explain the missing Taboo entries for Nearest Neighbors and Random Forest; this should be clarified, and the test-set class distribution should be reported.
[§3.2] The blinding of the annotators to the original/paraphrased status is not convincingly described. Each annotator's batch contains both the original sentence and its paraphrased version, and the paper only says that 'no explicit indication was provided regarding the sentence's pair.' Because the two sentences often share many lexical items, an annotator can easily infer which is the paraphrase, especially when one version is visibly more polite or neutral. This could subtly influence their sensitivity labels and inflate the measured direction of the shift. The authors should describe how the two versions of a pair were presented to the annotator (e.g., separated across batches, randomized order, or masked formatting) or otherwise justify that the pairing was not inferable.
[§4.1] The confusion matrix in Figure 2 is reported only as row percentages, without the underlying counts or the total number of paired sentences. The Bowker test statistic is reported, but the strength of the evidence cannot be fully assessed without knowing the number of observations per cell (for instance, whether the 87.1% diagonal for Formal/Polite is based on 50 or 500 sentences). Reporting the raw counts or at least the row sums would allow readers to evaluate the precision of the estimated transition probabilities and the robustness of the Bowker test result.

minor comments (5)

[§3.2] The category definition for Informal contains a typo: 'very conversetional' should be 'very conversational.'
[§3.3] The Δ metric assumes an equal-interval scale across the four sensitivity categories (1=Formal/Polite to 4=Taboo). This is a strong assumption, as the perceptual distance between Derogatory and Taboo may differ from that between Informal and Derogatory; the paper should acknowledge this limitation explicitly.
[§3.1] The paper reports 23,347 sentences collected but does not state how the 1,250 original sentences were sampled 'uniformly with respect to expressions.' Since there are 599 expressions, uniform sampling would give roughly 2 sentences per expression; the authors should clarify whether this was a random sample or a stratified selection, and how this interacts with the at-most-45-sentences-per-expression constraint.
[§4.2] The zero-shot LLM classification evaluation does not describe how the model outputs were parsed (e.g., whether outputs were normalized, whether synonyms like 'formal' or 'taboo' were accepted) or how failed outputs were handled. This information is needed to interpret the reported F1 scores.
[Table 1] The rows for Nearest Neighbors and Random Forest show a dash for the Taboo Δ value, which is unexplained in the text. If these classifiers never predict the Taboo class in the test set, that is an important limitation that should be noted in the discussion of the classifiers' performance.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the moderation claim is measured from expert annotations external to GPT-4o-mini, with no parameter fitted to the predicted shift.

full rationale

The derivation chain is: collect sentences by expression, paraphrase with GPT-4o-mini, have human experts assign one of four sensitivity labels, then compare original versus paraphrase labels via a confusion matrix, Bowker's test, and the Delta statistic. The expert annotations are external ground truth, not outputs of the model being tested, and no equation in Section 3.3 defines Delta in terms of the conclusion; Delta is simply the mean of (paraphrase label - original label). The central claim (GPT-4o-mini systematically reduces sensitivity) is an empirical reading of those measured differences, not an identity with any input. The open-source LLM and traditional classifier experiments are standard benchmark evaluations against the same expert labels; training them on 80% of the labels and testing on 20% is not fitted-input-called-prediction. The only self-citation is Born Classifier (Guidotti and Ferrara, 2022), used as a comparative baseline in Tables 1 and 2; it is a published, externally checkable method and is not load-bearing for the moderation finding, so it does not raise the circularity score. Concerns about inter-annotator agreement, adjudication rules, or the 231 unaccounted sentences are validity/measurement-quality issues, not circularity.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The central measurement is empirical and has no fitted free parameters; the main burdens are the validity of the ordinal sensitivity scale, the reliability of expert labels, and the neutrality of the paraphrase instruction.

assumptions (3)

domain assumption The four sensitivity classes form an ordinal scale where Formal/Polite < Informal < Derogatory < Taboo and the numeric differences between adjacent classes are meaningfully comparable.
Section 3.2 defines the classes and Section 3.3 treats them as equally spaced levels 1-4 for the Delta metric; the interpretation of a shift of one or two levels assumes this comparability.
domain assumption Expert annotations are assumed to be valid ground truth for sentence sensitivity, including the implicit assumption that two independent annotations without reported agreement or aggregation rule are sufficient.
Section 3.2 assigns two annotations per sentence but does not report agreement or the aggregation rule; Section 4 treats these labels as ground truth for both Delta and classifier accuracy.
ad hoc to paper The prompt 'Paraphrase the following sentence: {original_sentence}' is a neutral instruction, so the observed shift is a property of the model's implicit moderation rather than of the paraphrase task itself.
Section 3.1: no control conditions, such as a style-preserving instruction or an unaligned model, are tested. This assumption is load-bearing for the interpretation of implicit moderation.

how reviews work

0 comments

Cite this review

Pith. "Pith review of What's Taboo for You? - An Empirical Evaluation of LLMs Behavior Toward Sensitive Content." pith.science (2026). https://pith.science/paper/5KAAAYIG

@misc{pith2026250723319,
  author       = {Pith},
  title        = {Pith review of: What's Taboo for You? - An Empirical Evaluation of LLMs Behavior Toward Sensitive Content},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5KAAAYIG}},
  note         = {Machine review of arXiv:2507.23319}
}

read the original abstract

Proprietary Large Language Models (LLMs) have shown tendencies toward politeness, formality, and implicit content moderation. While previous research has primarily focused on explicitly training models to moderate and detoxify sensitive content, there has been limited exploration of whether LLMs implicitly sanitize language without explicit instructions. This study empirically analyzes the implicit moderation behavior of GPT-4o-mini when paraphrasing sensitive content and evaluates the extent of sensitivity shifts. Our experiments indicate that GPT-4o-mini systematically moderates content toward less sensitive classes, with substantial reductions in derogatory and taboo language. Also, we evaluate the zero-shot capabilities of LLMs in classifying sentence sensitivity, comparing their performances against traditional methods.

Figures

Figures reproduced from arXiv: 2507.23319 by the authors.

**Figure 2.** The confusion matrix shows sensitivity classes [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

**Figure 3.** Confusion matrices show sensitivity classifications (Original vs. Paraphrased) for Open-Source LLMs [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 20 canonical work pages

[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]

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]

Meysam Alizadeh, Ma \"e l Kubli, Zeynab Samei, Shirin Dehghani, Mohammadmasiha Zahedivafa, Juan D Bermeo, Maria Korobeynikova, and Fabrizio Gilardi. 2025. Open-source LLMs for text annotation: a practical guide for model setting and fine-tuning. Journal of Computational Social Science, 8(1):17

work page 2025
[4]

John Ayto, editor. 1999. Oxford Dictionary of Slang, 2 edition. Oxford University Press, London, England

work page 1999
[5]

Bowman, Zac Hatfield-Dodds, Ben Mann, Dario Amodei, Nicholas Joseph, Sam McCandlish, Tom Brown, and Jared Kaplan

Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, Carol Chen, Catherine Olsson, Christopher Olah, Danny Hernandez, Dawn Drain, Deep Ganguli, Dustin Li, Eli Tran-Johnson, Ethan Perez, Jamie Kerr, Jared Mueller, Jeffrey Ladish, Joshua Landau, Kamal Ndousse, K...

arXiv 2022
[6]

A H Bowker. 1948. A test for symmetry in contingency tables. Journal of the American Statistical Association, 43(244):572--574

work page 1948
[7]

Leo Breiman. 2001. Random forests. Machine Learning, 45(1):5--32

work page 2001
[8]

Corinna Cortes and Vladimir Vapnik. 1995. Support-vector networks. Machine Learning, 20(3):273--297

work page 1995

Show all 38 references

[9]

Thomas Cover and Peter E Hart. 1967. https://doi.org/10.1109/TIT.1967.1053964 Nearest neighbor pattern classification . IEEE Transactions on Information Theory, 13(1):21--27

1967
[10]

David Dale, Anton Voronov, Daryna Dementieva, Varvara Logacheva, Olga Kozlova, Nikita Semenov, and Alexander Panchenko. 2021. https://doi.org/10.18653/v1/2021.emnlp-main.629 Text detoxification using large pre-trained neural models . In Proceedings of the 2021 Conference on Em...

2021 doi
[11]

Tom Dalzell, editor. 2018. The Routledge dictionary of modern American slang and unconventional English, 2 edition. Routledge, London, England

2018
[12]

Daryna Dementieva, Daniil Moskovskiy, Nikolay Babakov, Abinew Ali Ayele, Naquee Rizwan, Florian Schneider, Xintong Wang, Seid Muhie Yimam, Dmitry Ustalov, Elisei Stakovskii, Alisa Smirnova, Ashraf Elnagar, Animesh Mukherjee, and Alexander Panchenko. 2024. https://ceur-ws.org/V...

2024
[13]

Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. https://doi.org/10.18653/v1/N19-1423 BERT : Pre-training of deep bidirectional transformers for language understanding . In Proceedings of the 2019 Conference of the North A merican Chapter of the Associat...

2019 doi
[14]

Samuel Gehman, Suchin Gururangan, Maarten Sap, Yejin Choi, and Noah A. Smith. 2020. https://doi.org/10.18653/v1/2020.findings-emnlp.301 R eal T oxicity P rompts: Evaluating neural toxic degeneration in language models . In Findings of the Association for Computational Linguist...

2020 doi
[15]

Gemma Team . 2024. https://arxiv.org/abs/2408.00118 Gemma 2: Improving open language models at a practical size . Preprint, arXiv:2408.00118

2024 arXiv
[16]

Fabrizio Gilardi, Meysam Alizadeh, and Maël Kubli. 2023. https://doi.org/10.1073/pnas.2305016120 Chatgpt outperforms crowd workers for text-annotation tasks . Proceedings of the National Academy of Sciences, 120(30):e2305016120

2023 doi
[17]

Emanuele Guidotti and Alfio Ferrara. 2022. https://proceedings.neurips.cc/paper_files/paper/2022/file/c88d0c9bea6230b518ce71268c8e49e0-Paper-Conference.pdf Text classification with born s rule . In Advances in Neural Information Processing Systems, volume 35, pages 30990--3100...

2022
[18]

David J Hand and Keming Yu. 2001. Idiot's bayes---not so stupid after all? International Statistical Review, 69(3):385--398

2001
[19]

Miloš Jakubíček, Adam Kilgarriff, Vojtěch Kovář, Pavel Rychlý, and Vít Suchomel. 2013. http://ucrel.lancs.ac.uk/cl2013/ The tenten corpus family . In 7th International Corpus Linguistics Conference CL 2013, pages 125--127, Lancaster

2013
[20]

Deepak Kumar, Yousef Anees AbuHashem, and Zakir Durumeric. 2024 a . Watch your language: Investigating content moderation with large language models. Proceedings of the International AAAI Conference on Web and Social Media, 18:865--878

2024
[21]

Deepak Kumar, Yousef Anees AbuHashem, and Zakir Durumeric. 2024 b . https://doi.org/10.1609/icwsm.v18i1.31358 Watch your language: Investigating content moderation with large language models . Proceedings of the International AAAI Conference on Web and Social Media, 18(1):865--878

2024 doi
[22]

Jiaxi Liu. 2024. ChatGPT : perspectives from human-computer interaction and psychology. Frontiers in Artificial Intelligence, 7:1418869

2024
[23]

Llama Team . 2024. https://arxiv.org/abs/2407.21783 The llama 3 herd of models . Preprint, arXiv:2407.21783

2024 arXiv
[24]

Aman Madaan, Amrith Setlur, Tanmay Parekh, Barnabas Poczos, Graham Neubig, Yiming Yang, Ruslan Salakhutdinov, Alan W Black, and Shrimai Prabhumoye. 2020. https://doi.org/10.18653/v1/2020.acl-main.169 Politeness transfer: A tag and generate approach . In Proceedings of the 58th...

2020 doi
[25]

Microsoft . 2024. https://arxiv.org/abs/2404.14219 Phi-3 technical report: A highly capable language model locally on your phone . Preprint, arXiv:2404.14219

2024 arXiv
[26]

Mistral AI Team . 2024. U n M inistral, des M inistraux. https://mistral.ai/news/ministraux. [Accessed 29-01-2024]

2024
[27]

Tong Niu and Mohit Bansal. 2018. https://doi.org/10.1162/tacl_a_00027 Polite dialogue generation without parallel data . Transactions of the Association for Computational Linguistics, 6:373--389

2018 doi
[28]

OpenAI. 2024 a . https://arxiv.org/abs/2303.08774 Gpt-4 technical report . Preprint, arXiv:2303.08774

2024 arXiv
[29]

OpenAI. 2024 b . G P T -4o mini: advancing cost-efficient intelligence. https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence. [Accessed 23-07-2024]

2024
[30]

Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul F Christiano, Jan Leike, a...

2022
[31]

Qwen Team . 2025. https://arxiv.org/abs/2412.15115 Qwen2.5 technical report . Preprint, arXiv:2412.15115

2025 arXiv
[32]

Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. 2019. Language models are unsupervised multitask learners. OpenAI blog, 1(8):9

2019
[33]

Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 21(1)

2020
[34]

David E Rumelhart, Geoffrey E Hinton, and Ronald J Williams. 1986. Learning representations by back-propagating errors. Nature, 323(6088):533--536

1986
[35]

Anirudh Som, Karan Sikka, Helen Gent, Ajay Divakaran, Andreas Kathol, and Dimitra Vergyri. 2024. https://doi.org/10.18653/v1/2024.findings-acl.749 Demonstrations are all you need: Advancing offensive content paraphrasing using in-context learning . In Findings of the Associati...

2024 doi
[36]

Richard A Spears. 1990. Forbidden American English. McGraw-Hill Contemporary, Maidenhead, England

1990
[37]

Team OLMo . 2025. https://arxiv.org/abs/2501.00656 2 olmo 2 furious . Preprint, arXiv:2501.00656

2025 arXiv
[38]

Emmanuel Vargas Penagos. 2024. ChatGPT , can you solve the content moderation dilemma? International Journal of Law and Information Technology, 32(1)

2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[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]

Meysam Alizadeh, Ma \"e l Kubli, Zeynab Samei, Shirin Dehghani, Mohammadmasiha Zahedivafa, Juan D Bermeo, Maria Korobeynikova, and Fabrizio Gilardi. 2025. Open-source LLMs for text annotation: a practical guide for model setting and fine-tuning. Journal of Computational Social Science, 8(1):17

work page 2025

[4] [4]

John Ayto, editor. 1999. Oxford Dictionary of Slang, 2 edition. Oxford University Press, London, England

work page 1999

[5] [5]

Bowman, Zac Hatfield-Dodds, Ben Mann, Dario Amodei, Nicholas Joseph, Sam McCandlish, Tom Brown, and Jared Kaplan

Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, Carol Chen, Catherine Olsson, Christopher Olah, Danny Hernandez, Dawn Drain, Deep Ganguli, Dustin Li, Eli Tran-Johnson, Ethan Perez, Jamie Kerr, Jared Mueller, Jeffrey Ladish, Joshua Landau, Kamal Ndousse, K...

arXiv 2022

[6] [6]

A H Bowker. 1948. A test for symmetry in contingency tables. Journal of the American Statistical Association, 43(244):572--574

work page 1948

[7] [7]

Leo Breiman. 2001. Random forests. Machine Learning, 45(1):5--32

work page 2001

[8] [8]

Corinna Cortes and Vladimir Vapnik. 1995. Support-vector networks. Machine Learning, 20(3):273--297

work page 1995