Comparison of Modern Multilingual Text Embedding Techniques for Hate Speech Detection Task

• URL: http://arxiv.org/abs/2604.14907v1
• Date: Thu, 16 Apr 2026 11:49:13 GMT
• Title: Comparison of Modern Multilingual Text Embedding Techniques for Hate Speech Detection Task
• Abstract summary: We introduce LtHate, a new Lithuanian hate speech corpus derived from news portals and social networks.<n>We benchmark six modern multilingual encoders (potion, gemma, bge, snow, jina, e5) on LtHate, RuToxic, and EnSuperset using a unified Python pipeline.<n>For each embedding, we train both a one class HBOS anomaly detector and a two class CatBoost, with and without principal component analysis (PCA) compression to 64-dimensional feature vectors.<n>Across all datasets, two class supervised models consistently and substantially outperform one class anomaly detection.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Online hate speech and abusive language pose a growing challenge for content moderation, especially in multilingual settings and for low-resource languages such as Lithuanian. This paper investigates to what extent modern multilingual sentence embedding models can support accurate hate speech detection in Lithuanian, Russian, and English, and how their performance depends on downstream modeling choices and feature dimensionality. We introduce LtHate, a new Lithuanian hate speech corpus derived from news portals and social networks, and benchmark six modern multilingual encoders (potion, gemma, bge, snow, jina, e5) on LtHate, RuToxic, and EnSuperset using a unified Python pipeline. For each embedding, we train both a one class HBOS anomaly detector and a two class CatBoost classifier, with and without principal component analysis (PCA) compression to 64-dimensional feature vectors. Across all datasets, two class supervised models consistently and substantially outperform one class anomaly detection, with the best configurations achieving up to 80.96% accuracy and AUC ROC of 0.887 in Lithuanian (jina), 92.19% accuracy and AUC ROC of 0.978 in Russian (e5), and 77.21% accuracy and AUC ROC of 0.859 in English (e5 with PCA). PCA compression preserves almost all discriminative power in the supervised setting, while showing some negative impact for the unsupervised anomaly detection case. These results demonstrate how modern multilingual sentence embeddings combined with gradient boosted decision trees provide robust soft-computing solutions for multilingual hate speech detection applications.

Related papers

• Beyond Cross-Lingual Transfer: Benchmarking Propagation Boundaries in Multilingual LLM Unlearning [80.05116806217397]
  Large Language Model (LLM) unlearning aims to suppress target knowledge while preserving general capabilities.<n>We introduce CLLPU (Cross-Lingual and Language-Bound Protocol for LLM Unlearning), a multilingual benchmark that formulates this problem through two settings.<n>Common-goal forgetting, where target knowledge should be suppressed across all languages, and language-conditioned forgetting, where suppression should remain confined to a designated language.
  arXiv  Detail & Related papers  (2026-09-05T08:26:15Z)
• Efficient Adaptation of LLMs for Hate Speech Detection in Low-Resource Languages: A Comparative Study on Roman Urdu [0.3436215734071903]
  It is challenging to detect hate speech in Low Resource Languages (LRLs) because of the absence of annotated data, the informality of its language structure, and the lack of standardized grammar.<n>The objective of this paper is to conduct a comprehensive assessment of Large Language Models (LLMs) for Hate Speech Detection (HSD) in Roman Urdu script.<n>Experiments are conducted on the PURUTT (Parallel Urdu and Roman Urdu Corpus for Toxic Comments and Transliteration) dataset with over 72,000 annotated comments.
  arXiv  Detail & Related papers  (2026-08-06T11:28:19Z)
• M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models [0.0]
  M-GATE is a benchmark of linguistic proficiency spanning 30 typologically diverse languages.<n>We evaluate over 50 models in more than 80 configurations.
  arXiv  Detail & Related papers  (2026-08-04T15:16:05Z)
• Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech [61.759910921200834]
  Cross-lingual sentence encoders typically cover only a few hundred languages.<n>We introduce OmniSONAR, a new family of omnilingual, cross-lingual and cross-language sentence embedding models.
  arXiv  Detail & Related papers  (2026-03-17T14:47:35Z)
• Multilingual Hate Speech Detection in Social Media Using Translation-Based Approaches with Large Language Models [4.66584517664999]
  We introduce a trilingual dataset of 10,193 tweets in English, Urdu, and Spanish, collected via keyword filtering.<n>Our approach, integrating attention layers with GPT-3.5 Turbo and Qwen 2.5 72B, achieves strong performance.<n>Our framework offers a robust solution for multilingual hate speech detection, fostering safer digital communities worldwide.
  arXiv  Detail & Related papers  (2025-06-09T18:53:56Z)
• Can Prompting LLMs Unlock Hate Speech Detection across Languages? A Zero-shot and Few-shot Study [59.30098850050971]
  This work evaluates LLM prompting-based detection across eight non-English languages.<n>We show that while zero-shot and few-shot prompting lag behind fine-tuned encoder models on most of the real-world evaluation sets, they achieve better generalization on functional tests for hate speech detection.
  arXiv  Detail & Related papers  (2025-05-09T16:00:01Z)
• Evaluation of Hate Speech Detection Using Large Language Models and Geographical Contextualization [0.0]
  This study systematically investigates the performance of LLMs on detecting hate speech across multilingual and diverse geographic contexts.<n>We evaluate three state-of-the-art LLMs: Llama2 (13b), Codellama (7b), and DeepSeekCoder (6.7b)<n>Codellama had the best binary classification recall with 70.6% and an F1-score of 52.18%, whereas DeepSeekCoder had the best performance in geographic sensitivity, correctly detecting 63 out of 265 locations.
  arXiv  Detail & Related papers  (2025-02-26T22:59:36Z)
• Code-Mixed Telugu-English Hate Speech Detection [0.0]
  This study investigates transformer-based models, including TeluguHateBERT, HateBERT, DeBERTa, Muril, IndicBERT, Roberta, and Hindi-Abusive-MuRIL, for classifying hate speech in Telugu.<n>We fine-tune these models using Low-Rank Adaptation (LoRA) to optimize efficiency and performance.<n>We translate Telugu text into English using Google Translate to assess its impact on classification accuracy.
  arXiv  Detail & Related papers  (2025-02-15T02:03:13Z)
• HateTinyLLM : Hate Speech Detection Using Tiny Large Language Models [0.0]
  Hate speech encompasses verbal, written, or behavioral communication that targets derogatory or discriminatory language against individuals or groups. HateTinyLLM is a novel framework based on fine-tuned decoder-only tiny large language models (tinyLLMs) for efficient hate speech detection.
  arXiv  Detail & Related papers  (2024-04-26T05:29:35Z)
• Robustification of Multilingual Language Models to Real-world Noise with Robust Contrastive Pretraining [14.087882550564169]
  We assess the robustness of neural models on noisy data and suggest improvements are limited to the English language. To benchmark the performance of pretrained multilingual models, we construct noisy datasets covering five languages and four NLP tasks. We propose Robust Contrastive Pretraining (RCP) to boost the zero-shot cross-lingual robustness of multilingual pretrained models.
  arXiv  Detail & Related papers  (2022-10-10T15:40:43Z)
• OneAligner: Zero-shot Cross-lingual Transfer with One Rich-Resource Language Pair for Low-Resource Sentence Retrieval [91.76575626229824]
  We present OneAligner, an alignment model specially designed for sentence retrieval tasks. When trained with all language pairs of a large-scale parallel multilingual corpus (OPUS-100), this model achieves the state-of-the-art result. We conclude through empirical results and analyses that the performance of the sentence alignment task depends mostly on the monolingual and parallel data size.
  arXiv  Detail & Related papers  (2022-05-17T19:52:42Z)
• Addressing the Challenges of Cross-Lingual Hate Speech Detection [115.1352779982269]
  In this paper we focus on cross-lingual transfer learning to support hate speech detection in low-resource languages. We leverage cross-lingual word embeddings to train our neural network systems on the source language and apply it to the target language. We investigate the issue of label imbalance of hate speech datasets, since the high ratio of non-hate examples compared to hate examples often leads to low model performance.
  arXiv  Detail & Related papers  (2022-01-15T20:48:14Z)
• Unsupervised Cross-lingual Representation Learning for Speech Recognition [63.85924123692923]
  XLSR learns cross-lingual speech representations by pretraining a single model from the raw waveform of speech in multiple languages. We build on wav2vec 2.0 which is trained by solving a contrastive task over masked latent speech representations. Experiments show that cross-lingual pretraining significantly outperforms monolingual pretraining.
  arXiv  Detail & Related papers  (2020-06-24T18:25:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.