Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

Enhancing LLM-based Hatred and Toxicity Detection with Meta-Toxic Knowledge Graph

T0 review · 4 major / 4 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 Injecting toxic-knowledge triplets into LLM prompts reduces false positives in hate speech detection.

desk verdict A genuinely new training-free KG-based toxicity detection method, but its headline FPR gains are confounded by a prompt bias and at least one suspicious duplicated table value. read the letter →

arxiv 2412.15268 v4 pith:L7FMDQUZ submitted 2024-12-17 cs.CL cs.AI

Yibo Zhao , Jiapeng Zhu , Can Xu , Yao Liu , Xiang Li This is my paper · ORCID
classification cs.CLcs.AI
keywords hatespeechdetectiontoxicityknowledgegraphretrieval-augmentedgenerationfalsepositiveslargelanguagemodelsretrievalentityresolution
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

MetaTox proposes to reduce two failure modes of LLM-based toxicity detection: false negatives from missing domain-specific knowledge, and false positives from over-sensitivity to identity-related words. It builds a meta-toxic knowledge graph from toxic examples in three existing benchmarks, distilling the examples into Subject-Predicate-Object triplets such as "white lives matter is against black lives matter." At detection time, the system extracts entities from the new text, maps them to graph nodes, retrieves shortest paths, ranks and filters the resulting triplets by similarity, and places the survivors in the LLM prompt. The paper reports consistently lower false positive rates and higher accuracy and F1 relative to vanilla prompting and naive RAG, with the largest gains in cross-domain evaluation. If correct, this gives a training-free route to updating and specializing toxicity detectors, while the in-domain results caution that fine-tuned small models remain strong competitors.

What carries the argument

The central object is the meta-toxic knowledge graph: a graph whose nodes are toxic entities and concepts and whose edges are relations such as "demean," "promote," and "co-appear," with each path standing for a piece of concentrated toxicity knowledge. Its role in the argument is to turn fragmented toxic corpus text into concise, queryable triplets that can be ranked against an incoming speech and placed into the LLM prompt. Two later mechanisms carry much of the reported gain: shortest-path retrieval between mapped entities, and a ranking-and-filtering step that discards triplets with low BERT-embedding similarity to the input, which the ablation study says is essential.

What would settle it

Run MetaTox on a held-out dataset while sweeping the ranking-and-filtering similarity threshold; if the optimal threshold changes across datasets or performance collapses when the threshold is fixed, the ranking mechanism is not stable. A simpler check is to have human annotators rate whether the retrieved triplets are actually relevant to the input speech and compare that rating with the cosine-similarity order.

Watch

Extended reading notes

Core claim

On its own terms, the paper claims that a knowledge graph of toxic triplets can carry the domain knowledge LLMs lack, and that retrieving from it makes the LLM's judgment both more sensitive to implicit hate and less allergic to benign mentions of sensitive groups. The construction pipeline treats LLMs as extractors rather than judges: given a known-toxic text, the LLM first writes a rationale, then extracts toxic triplets, then filters them with a self-checking step, and entity resolution merges near-duplicate nodes using BERT embeddings. At query time, entity extraction, node mapping, shortest-path retrieval, and similarity-based ranking filter the candidate triplets, and the prompt tells the model that a speech matching the triplets is hateful while a speech with no related triplets is likely benign. This design leads to the paper's headline results: false positive rate falls sharply on both HateXplain and IHC—for instance from 66.62% to 32.10% with Qwen on HateXplain—while accuracy and F1 rise, and cross-domain transfers from a ToxicSpans-built graph still beat both vanilla LLM prompting and naive RAG.

Load-bearing premise

The method assumes that BERT-embedding cosine similarity between a candidate triplet and the incoming speech reliably indicates whether that toxic knowledge is relevant, and the paper does not specify or separately validate the similarity threshold used to filter triplets.

Editorial extensions

If this is right

  • On HateXplain and IHC, MetaTox raises accuracy, F1, and AUC relative to vanilla LLM prompting and naive RAG while cutting the false positive rate, for example from 66.62% to 32.10% with Qwen on HateXplain.
  • Because the method is training-free, the knowledge source can be expanded or edited without fine-tuning, which suits fast-evolving social media toxicity.
  • Using a graph built from ToxicSpans to test on HateXplain and IHC, MetaTox keeps much of its gain while naive RAG degrades, supporting cross-domain transfer of invariant toxic semantics.
  • In-domain results suggest fine-tuned small models such as HateBERT and BERT remain competitive or superior, so model choice depends on whether the deployment domain matches the training domain.
  • Ablation results show that both the ranking/filtering step and shortest-path retrieval, rather than 1-hop neighbor retrieval, contribute to the reported accuracy.

Reading between the lines

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

  • The ranking step's reliance on unspecified BERT-embedding similarity thresholds is the least controlled part of the pipeline; if the threshold was tuned separately per test set, the reported false-positive reductions may not transfer to new corpora.
  • The prompt instruction that no related triplets implies non-toxicity could bias the model toward benign predictions, so part of the false-positive reduction may come from instruction bias rather than knowledge quality.
  • Manual evaluation reports 82–89% triplet correctness, so the graph carries residual noise; measuring how detection performance changes as triplet accuracy is degraded artificially would test whether correctness is the active ingredient.
  • A natural extension is to build the graph from unlabeled social media streams rather than curated benchmarks and evaluate whether retrieval still improves detection, which would test the method's claim to be easily updateable.
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 / 4 minor

Summary. The paper proposes MetaTox, a training-free method for LLM-based hatred and toxicity detection. It constructs a meta-toxic knowledge graph from toxic training samples of HateXplain, ToxicSpans, and IHC using a three-step pipeline (rationale reasoning, triplet extraction, entity resolution). At inference, the method extracts entities from a given text, maps them to graph nodes, retrieves shortest paths, and injects ranked toxic triplets into the LLM prompt. Experiments compare MetaTox against vanilla LLM, naive RAG, and fine-tuned small models in both in-domain and cross-domain settings, reporting accuracy, F1, AUC, and false positive rate (FPR). The central claim is that MetaTox significantly reduces false positives while improving overall detection performance, especially out-of-domain.

Significance. If the causal claim holds, MetaTox offers a practical, training-free way to inject domain-specific toxicity knowledge into LLMs, with a novel domain-specific knowledge graph as a reusable artifact. The paper merits credit for releasing code, performing a manual annotation-quality check of the constructed graphs, including an ablation study, and evaluating cross-domain transfer, which is a meaningful step beyond in-domain-only comparisons. However, the current experimental design does not isolate the effect of the retrieved triplets from the effect of a strong prompt prior, and the absence of thresholds and statistical reporting leaves the headline FPR reductions unsubstantiated.

major comments (4)
  1. [Appendix A.5 / Section 3.3] The query prompt in Appendix A.5 instructs the model: 'If the context contains the same meaning with one of the triplets, it is hateful. Otherwise, it is more possible to be benign, but you should leverage your knowledge to judge it.' This is a strong default-to-benign prior that is not present in the vanilla or naive-RAG baselines as described. Because the ablation in Appendix B only removes rank-and-filter while retaining this instruction, the reported FPR reductions (e.g., Table 2, HateXplain with Qwen: 66.62 to 32.10; Table 3, IHC with Qwen: 48.42 to 11.64) cannot be attributed to the injected knowledge graph triplets. The authors must report a controlled condition with the same prompt instruction but (a) no triplets, and (b) randomly sampled irrelevant triplets, to separate the prompt effect from the knowledge effect.
  2. [Tables 2 and 3] Several FPR values are exactly identical between the in-domain and cross-domain tables for the same backbone and test dataset: HateXplain with Qwen shows MetaTox FPR 32.10 in both Table 2 (KG from same dataset) and Table 3 (KG from ToxicSpans), and IHC with Llama shows 34.23 in both tables. Since the knowledge graph source differs across the two settings, exact duplication is implausible and suggests an experiment or reporting error. The authors must verify these numbers and explain the duplication or correct the tables.
  3. [Section 3.3 and Section 3.2.3] The similarity threshold used in 'Ranking and Filtering' is never specified, and the entity-resolution clustering threshold is described only as 'relatively high' with no numerical value. These thresholds directly control how many triplets are injected into the prompt and hence how often the default-to-benign instruction applies. Without reporting the thresholds and a sensitivity analysis, the results are not reproducible and the contribution of the filtering step cannot be assessed.
  4. [Section 4.2] All experiments in Tables 2 and 3 report single runs with no error bars, confidence intervals, or significance tests. The abstract's claim that MetaTox 'significantly decreases the false positive rate' is therefore not statistically supported. At minimum, the authors should provide multiple seeds or bootstrap confidence intervals for the headline FPR and AUC differences.
minor comments (4)
  1. [Appendix A.5] The query prompt contains typos: 'aplphabet' should be 'alphabet' and 'Non-hatefule' should be 'Non-hateful'.
  2. [Appendix B / Appendix C] The appendix text for Section B and Section C appears duplicated in the manuscript, and the self-checking prompt in Section A.2 is printed twice; this likely stems from a formatting error and should be corrected.
  3. [Table 2] The column header 'RAGMetaTox' is missing a space; it should read 'RAG MetaTox' to match the other method names.
  4. [Section 3.1] The knowledge graph is constructed from the same training corpora used by the naive-RAG baseline in the in-domain setting; while this is a fair comparison, the paper does not discuss the extent to which in-domain gains could reflect the LLM's familiarity with those exact training texts. A brief discussion of this circularity concern would strengthen the analysis.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the knowledge graph is built from training splits and evaluated on held-out test splits, with a cross-domain setting that breaks any memorization loop.

full rationale

MetaTox's central claim is empirical: the authors build a KG from toxic training samples (Section 3.1), retrieve relevant triplets for a given input (Section 3.3), and inject them into an LLM prompt (Appendix A.5). The KG is constructed exclusively from the training splits of HateXplain, ToxicSpans, and IHC, while evaluation is on the corresponding test splits. Critically, the cross-domain experiments build both the KG and the naive-RAG corpus from ToxicSpans but evaluate on HateXplain and IHC (Table 3), so the reported FPR improvements cannot be explained by retrieval of memorized test content. No fitted parameter is renamed as a prediction: the BERT-embedding similarity thresholds used in entity resolution and ranking are design choices, and the paper does not present them as tuned on test labels. The query prompt's instruction that 'Otherwise, it is more possible to be benign' (Appendix A.5) is a plausible alternative explanation for the FPR reduction, but it is a confound in causal attribution, not a circular derivation: the model's output is not definitionally equal to the prompt instruction or to the retrieved triplets. The ablation in Appendix B varies retrieval components while keeping the prompt, so the paper does not reduce its central claim to a self-consistency equation. There are no load-bearing self-citations; citations to STaR, GraphRAG, and prior work on LLM sensitivity are external. Under the requirement to exhibit a specific reduction of a reported result to its own inputs, no circular step is established.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the reliability of LLM extraction, the representativeness of three benchmark datasets, and the validity of embedding-based similarity for retrieval, plus two undisclosed similarity thresholds. The constructed knowledge graphs are not currently released.

free parameters (3)
  • Entity resolution clustering threshold
    Section 3.2.3 says 'we set a relatively high similarity threshold' but gives no value; determines whether entities like 'Jew' and 'jews' merge, directly affecting graph structure and retrieval.
  • Ranking and filtering similarity threshold
    Section 3.3 describes filtering 'less related knowledge based on similarity'; no threshold value is given, and it is not stated whether it was set on a dev set or the test sets.
  • Retrieved knowledge count
    The number of triplets injected into the prompt is not fixed; the paper says 'each remaining knowledge will ultimately be sent to the LLM' after filtering, so the effective prompt length is a free choice.
assumptions (4)
  • domain assumption The three benchmark datasets (HateXplain, ToxicSpans, IHC) provide reliable ground-truth toxicity labels and are representative of toxic speech.
    Used as the sole source of toxic knowledge and as evaluation sets (Section 3.1).
  • domain assumption Qwen2.5-14B-Instruct generates accurate rationales, triplets, and self-checking judgments for toxic content.
    The entire graph is built by this model; no independent verification of extraction quality beyond a 100-triplet manual sample per dataset (Section 4.1).
  • standard math BERT sentence embeddings capture the semantic similarity needed for entity merging and node mapping.
    Entity resolution (Section 3.2.3) and node mapping (Section 3.3) rely on pre-trained BERT embeddings; no probe of embedding quality is provided.
  • ad hoc to paper Cosine similarity between a candidate triplet and the input speech is a valid relevance signal for knowledge ranking.
    Used in Ranking and Filtering (Section 3.3); this is the paper's own choice and is not validated against any gold standard, and the threshold is undisclosed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Enhancing LLM-based Hatred and Toxicity Detection with Meta-Toxic Knowledge Graph." pith.science (2026). https://pith.science/paper/L7FMDQUZ

@misc{pith2026241215268,
  author       = {Pith},
  title        = {Pith review of: Enhancing LLM-based Hatred and Toxicity Detection with Meta-Toxic Knowledge Graph},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L7FMDQUZ}},
  note         = {Machine review of arXiv:2412.15268}
}
read the original abstract

The rapid growth of social media platforms has raised significant concerns regarding online content toxicity. When Large Language Models (LLMs) are used for toxicity detection, two key challenges emerge: 1) the absence of domain-specific toxic knowledge leads to false negatives; 2) the excessive sensitivity of LLMs to toxic speech results in false positives, limiting freedom of speech. To address these issues, we propose a novel method called MetaTox, leveraging graph search on a meta-toxic knowledge graph to enhance hatred and toxicity detection. First, we construct a comprehensive meta-toxic knowledge graph by utilizing LLMs to extract toxic information through a three-step pipeline, with toxic benchmark datasets serving as corpora. Second, we query the graph via retrieval and ranking processes to supplement accurate, relevant toxic knowledge. Extensive experiments and in-depth case studies across multiple datasets demonstrate that our MetaTox significantly decreases the false positive rate while boosting overall toxicity detection performance. Our code is available at https://github.com/YiboZhao624/MetaTox.

Figures

Figures reproduced from arXiv: 2412.15268 by the authors.

Figure 1
Figure 1. A showcase of our method MetaTox, which enhances the LLM to detect toxicity by injecting speech￾related toxic knowledge in form of triplets. critical task to detect the toxic content , while cau￾tiously avoiding hindering freedom of expression. To address the problem, previous work can be mainly categorized into three types: (1) rule-based methods (Chen et al. , 2012 ; Gitari et al. , 2015 ) relying on the pre-defin… view at source ↗
Figure 2
Figure 2. The pipeline of meta-toxic knowledge graph [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Graph query pipeline of MetaTox. We propose five steps including entity extraction, node mapping, path retrieval, formatting, and ranking and filtering to inject accurate knowledge to the LLM. volved. Hence, this rationale reasoning step can be interpreted as a form of data augmentation, which draws out implicit toxic elements from the speech and makes them more explicit to better guide the LLM’s reasoning logic tow… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Example of false negative misjudgment caused by lack of domain-specific knowledge. From top to bottom, each block shows the ground truth label and context, followed by the predictions and reasoning of the vanilla LLM, naive RAG, and MetaTox. training data (Chu et al., …

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Breaking the Cloak! Unveiling Chinese Cloaked Toxicity with Homophone Graph and Toxic Lexicon

    cs.CL 2025-05 conditional novelty 6.0 of 10

    C2TU combines a Chinese pronunciation graph, a toxic lexicon, and language-model probability checking to find and correct homophone-cloaked toxic words without any training.

Reference graph

Works this paper leans on

33 extracted references · 11 canonical work pages · cited by 1 Pith paper

  1. [1]

    AI@Meta. 2024. https://github.com/meta-llama/llama3/blob/main/MODEL_CARD.md Llama 3 model card

    2024

  2. [2]

    Dmitriy Bespalov, Sourav Bhabesh, Yi Xiang, Liutong Zhou, and Yanjun Qi. 2023. https://doi.org/10.18653/v1/2023.acl-industry.56 Towards building a robust toxicity predictor . In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track), pages 581--598, Toronto, Canada. Association for Computational ...

    doi:10.18653/v1/2023.acl-industry.56 2023

  3. [3]

    Tommaso Caselli, Valerio Basile, Jelena Mitrovi \'c , and Michael Granitzer. 2021. https://doi.org/10.18653/v1/2021.woah-1.3 H ate BERT : Retraining BERT for abusive language detection in E nglish . In Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021), pages 17--25, Online. Association for Computational Linguistics

    doi:10.18653/v1/2021.woah-1.3 2021

  4. [4]

    Ying Chen, Yilu Zhou, Sencun Zhu, and Heng Xu. 2012. Detecting offensive language in social media to protect adolescent online safety. In 2012 international conference on privacy, security, risk and trust and 2012 international confernece on social computing, pages 71--80. IEEE

    work page 2012

  5. [5]

    Le, Sergey Levine, and Yi Ma

    Tianzhe Chu, Yuexiang Zhai, Jihan Yang, Shengbang Tong, Saining Xie, Dale Schuurmans, Quoc V. Le, Sergey Levine, and Yi Ma. 2025. https://arxiv.org/abs/2501.17161 Sft memorizes, rl generalizes: A comparative study of foundation model post-training . Preprint, arXiv:2501.17161

    arXiv 2025

  6. [6]

    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 Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long a...

    doi:10.18653/v1/n19-1423 2019

  7. [7]

    Nemanja Djuric, Jing Zhou, Robin Morris, Mihajlo Grbovic, Vladan Radosavljevic, and Narayan Bhamidipati. 2015. Hate speech detection with comment embeddings. In Proceedings of the 24th international conference on world wide web, pages 29--30

    work page 2015

  8. [8]

    Matthijs Douze, Alexandr Guzhva, Chengqi Deng, Jeff Johnson, Gergely Szilvasy, Pierre-Emmanuel Mazaré, Maria Lomeli, Lucas Hosseini, and Hervé Jégou. 2024. https://arxiv.org/abs/2401.08281 The faiss library

    arXiv 2024

Show all 33 references
  1. [9]

    Organization of Knowledge and Advanced Technologies

    Ashwin Geet D'Sa, Irina Illina, and Dominique Fohr. 2020. https://doi.org/10.1109/OCTA49274.2020.9151853 Bert and fasttext embeddings for automatic detection of toxic speech . In 2020 International Multi-Conference on: “Organization of Knowledge and Advanced Technologies” (OCT...

    2020

  2. [10]

    Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, and Jonathan Larson. 2024. From local to global: A graph rag approach to query-focused summarization. arXiv preprint arXiv:2404.16130

    2024 arXiv

  3. [11]

    Mai ElSherief, Caleb Ziems, David Muchlinski, Vaishnavi Anupindi, Jordyn Seybolt, Munmun De Choudhury, and Diyi Yang. 2021. https://aclanthology.org/2021.emnlp-main.29 Latent hatred: A benchmark for understanding implicit hate speech . In Proceedings of the 2021 Conference on ...

    2021

  4. [12]

    Njagi Dennis Gitari, Zhang Zuping, Hanyurwimfura Damien, and Jun Long. 2015. A lexicon-based approach for hate speech detection. International Journal of Multimedia and Ubiquitous Engineering, 10(4):215--230

    2015

  5. [13]

    Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021. Measuring massive multitask language understanding. Proceedings of the International Conference on Learning Representations (ICLR)

    2021

  6. [14]

    Nayeon Lee, Chani Jung, Junho Myung, Jiho Jin, Jose Camacho-Collados, Juho Kim, and Alice Oh. 2024. https://doi.org/10.18653/v1/2024.naacl-long.236 Exploring cross-cultural differences in E nglish hate speech annotations: From dataset construction to analysis . In Proceedings ...

    2024 doi

  7. [15]

    u ttler, Mike Lewis, Wen-tau Yih, Tim Rockt\

    Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich K\" u ttler, Mike Lewis, Wen-tau Yih, Tim Rockt\" a schel, Sebastian Riedel, and Douwe Kiela. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. In Proc...

    2020

  8. [16]

    Shuhua Liu and Thomas Forss. 2015. New classification models for detecting hate and violence web content. In 2015 7th international joint conference on knowledge discovery, knowledge engineering and knowledge management (IC3K), volume 1, pages 487--495. IEEE

    2015

  9. [17]

    Son T Luu and Ngan Luu-Thuy Nguyen. 2021. Uit-ise-nlp at semeval-2021 task 5: Toxic spans detection with bilstm-crf and toxicbert comment classification. arXiv preprint arXiv:2104.10100

    2021 arXiv

  10. [18]

    Binny Mathew, Punyajoy Saha, Seid Muhie Yimam, Chris Biemann, Pawan Goyal, and Animesh Mukherjee. 2021. Hatexplain: A benchmark dataset for explainable hate speech detection. In Proceedings of the AAAI conference on artificial intelligence, volume 35, pages 14867--14875

    2021

  11. [19]

    John Morris, Eli Lifland, Jin Yong Yoo, Jake Grigsby, Di Jin, and Yanjun Qi. 2020. https://doi.org/10.18653/v1/2020.emnlp-demos.16 T ext A ttack: A framework for adversarial attacks, data augmentation, and adversarial training in NLP . In Proceedings of the 2020 Conference on ...

    2020 doi

  12. [20]

    Ji Ho Park and Pascale Fung. 2017. One-step and two-step classification for abusive language detection on twitter. arXiv preprint arXiv:1706.01206

    2017 arXiv

  13. [21]

    John Pavlopoulos, Jeffrey Sorensen, L \'e o Laugier, and Ion Androutsopoulos. 2021. Semeval-2021 task 5: Toxic spans detection. In Proceedings of the 15th international workshop on semantic evaluation (SemEval-2021), pages 59--69

    2021

  14. [22]

    Qwen Team . 2024. https://qwenlm.github.io/blog/qwen2.5/ Qwen2.5: A party of foundation models

    2024

  15. [23]

    Robert Mark Simpson. 2013. Dignity, harm, and hate speech. Law and Philosophy, 32(6):701--728

    2013

  16. [24]

    Robert Mark Simpson. 2019. ‘won’t somebody please think of the children?’hate speech, harm, and childhood. Law and Philosophy, 38(1):79--108

    2019

  17. [25]

    William Warner and Julia Hirschberg. 2012. Detecting hate speech on the world wide web. In Proceedings of the second workshop on language in social media, pages 19--26

    2012

  18. [26]

    Chi, Quoc V

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H. Chi, Quoc V. Le, and Denny Zhou. 2024. Chain-of-thought prompting elicits reasoning in large language models. In Proceedings of the 36th International Conference on Neural Information Processi...

    2024

  19. [27]

    Yongjin Yang, Joonkee Kim, Yujin Kim, Namgyu Ho, James Thorne, and Se-Young Yun. 2023. https://doi.org/10.18653/v1/2023.findings-emnlp.365 HARE : Explainable hate speech detection with step-by-step reasoning . In Findings of the Association for Computational Linguistics: EMNLP...

    2023 doi

  20. [28]

    Griffiths, Yuan Cao, and Karthik Narasimhan

    Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, and Karthik Narasimhan. 2024. Tree of thoughts: deliberate problem solving with large language models. In Proceedings of the 37th International Conference on Neural Information Processing Systems,...

    2024

  21. [29]

    Eric Zelikman, Yuhuai Wu, Jesse Mu, and Noah Goodman. 2022. https://proceedings.neurips.cc/paper_files/paper/2022/file/639a9a172c044fbb64175b5fad42e9a5-Paper-Conference.pdf Star: Bootstrapping reasoning with reasoning . In Advances in Neural Information Processing Systems, vol...

    2022

  22. [30]

    Jiang Zhang, Qiong Wu, Yiming Xu, Cheng Cao, Zheng Du, and Konstantinos Psounis. 2024 a . https://doi.org/10.1609/aaai.v38i19.30178 Efficient toxic content detection by bootstrapping and distilling large language models . Proceedings of the AAAI Conference on Artificial Intell...

    2024 doi

  23. [31]

    Min Zhang, Jianfeng He, Taoran Ji, and Chang-Tien Lu. 2024 b . https://doi.org/10.18653/v1/2024.acl-long.652 Don ' t go to extremes: Revealing the excessive sensitivity and calibration limitations of LLM s in implicit hate speech detection . In Proceedings of the 62nd Annual M...

    2024 doi

  24. [32]

    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...

  25. [33]

    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 gl...

Pith tools

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