UIT-E10dot3 at SemEval-2021 Task 5: Toxic Spans Detection with Named Entity Recognition and Question-Answering Approaches

Hoang, Phu Gia; Nguyễn, Luân; Nguyen, Kiet Van

doi:10.18653/v1/2021.semeval-1.125

Cited by 4 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Luu et al [166], ToxicBERT initially evaluates the model's overall toxicity score and subsequently employs a toxic token-level classification model. Meanwhile, in Gia Hoang et al [161], a single-span toxic detector is employed to extract toxic words until ToxicBERT no longer deems them as toxic. Similarly to developments in sentencelevel toxic detection, ensemble models are utilized to improve upon previous results in numerous articles [152,154,163,165,167,168,173].…”

Section: Toxic Span Detectionmentioning

confidence: 99%

“…BERT+CNN [135,144,198] BERT+CRF [62,109,110,173] BERT+LSTM+CRF [110,173,176] BiLSTM+CRF [62,66,152,166,174] Spacy NER [152,158,161] Multichannel Convolutional Bidirectional Gated Recurrent Unit (MCBiGRU) [258] [180]…”

Section: Model Name Referencedmentioning

confidence: 99%

“…XLM-RoBERTa [255] [82, 159, 169, 196] ELECTRA [259] [114, 121, 147, 154, 155, 169, 212] Linear Support Vector Machine [106] SpanBERT [260] [67, 67, 106, 158, 176] ToxicBERT [261] [ 161,166] ALBERT+BiLSTM+CRF [162] CharacterBERT [262] [163]…”

Section: Model Name Referencedmentioning

confidence: 99%

See 2 more Smart Citations

A Systematic Review of Toxicity in Large Language Models: Definitions, Datasets, Detectors, Detoxification Methods and Challenges

Villate-Castillo,

Ser,

Urquijo

2024

Preprint

View full text Add to dashboard Cite

The emergence of the transformer architecture has ushered in a new era of possibilities, showcasing remarkable capabilities in generative tasks exemplified by models like GPT4o, Claude 3, and Llama 3. However, these advancements come with a caveat: predominantly trained on data gleaned from social media platforms, these systems inadvertently perpetuate societal biases and toxicity. Recognizing the paramount importance of AI Safety and Alignment, our study embarks on a thorough exploration through a comprehensive literature review focused on toxic language. Delving into various definitions, detection methodologies, and mitigation strategies, we aim to shed light on the complexities of this issue. While our focus primarily centres on transformer-based architectures, we also acknowledge and incorporate existing research within the realm of deep learning. Through our investigation, we uncover a multitude of challenges inherent in toxicity mitigation and detection models. These challenges range from inherent biases and generalization issues to the necessity for standardized definitions of toxic language and the quality assurance of dataset annotations. Furthermore, we emphasize the significance of transparent annotation processes, resolution of annotation disagreements, and the enhancement of Large Language Models (LLMs) robustness. Additionally, we advocate for the creation of standardized benchmarks to gauge the effectiveness of toxicity mitigation and detection methods. Addressing these challenges is not just imperative, but pivotal in advancing the development of safer and more ethically aligned AI systems.

show abstract

Section: Toxic Span Detectionmentioning

confidence: 99%

Section: Model Name Referencedmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Review of Toxicity in Large Language Models: Definitions, Datasets, Detectors, Detoxification Methods and Challenges

Villate-Castillo,

Ser,

Urquijo

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The architecture of the work was also available in the HuggingFace Transformer library. This library benefited the research community and was incorporated by different studies to attain other tasks [132], [133].…”

Section: ) Toxicity Featuresmentioning

confidence: 99%

Cyberbullying Detection in Social Networks: A Comparison Between Machine Learning and Transfer Learning Approaches

Teng

Varathan

2023

IEEE Access

View full text Add to dashboard Cite

Information and Communication Technologies fueled social networking and facilitated communication. However, cyberbullying on the platform had detrimental ramifications. The user-dependent mechanisms like reporting, blocking, and removing bullying posts online is manual and ineffective. Bagof-words text representation without metadata limited cyberbullying post text classification. This research developed an automatic system for cyberbullying detection with two approaches: Conventional Machine Learning and Transfer Learning. This research adopted AMiCA data encompassing significant amount of cyberbullying context and structured annotation process. Textual, sentiment and emotional, static and contextual word embeddings, psycholinguistics, term lists, and toxicity features were used in the conventional Machine Learning approach. This study was the first to use toxicity features to detect cyberbullying. This study is also the first to use the latest psycholinguistics features from the Linguistic Inquiry and Word (LIWC) 2022 tool, as well as Empath's lexicon, to detect cyberbullying. The contextual embeddings of ggeluBert, tnBert, and DistilBert have alike performance, however DistilBert embeddings were elected for higher F-measure. Textual features, DistilBert embeddings, and toxicity features that struck new benchmark were the top three unique features when fed individually. The model's performance was boosted to F-measure of 64.8% after feeding with a combination of textual, sentiment, DistilBert embeddings, psycholinguistics, and toxicity features to the Logistic Regression model that outperforms Linear SVC with faster training time and efficient handling of high-dimensionality features. Transfer Learning approach was by fine-tuning optimized version Pre-trained Language Models namely, DistilBert, DistilRoBerta, and Electra-small which were found to have speedier training computation than their base form. The fine-tuned DistilBert resulted with the highest F-measure of 72.42%, surpassing CML. Our research concluded that Transfer Learning was the best for uplifted performance and lesser effort as feature engineering and resampling was omitted.

show abstract

Evaluation of Word Embeddings for Toxic Span Prediction

Nelatoori

Kommanti

2022

Proceedings of the International Conference on Cognitive and Intelligent Computing

View full text Add to dashboard Cite

UIT-E10dot3 at SemEval-2021 Task 5: Toxic Spans Detection with Named Entity Recognition and Question-Answering Approaches

Cited by 4 publications

References 14 publications

A Systematic Review of Toxicity in Large Language Models: Definitions, Datasets, Detectors, Detoxification Methods and Challenges

A Systematic Review of Toxicity in Large Language Models: Definitions, Datasets, Detectors, Detoxification Methods and Challenges

Cyberbullying Detection in Social Networks: A Comparison Between Machine Learning and Transfer Learning Approaches

Evaluation of Word Embeddings for Toxic Span Prediction

Contact Info

Product

Resources

About