Papers › Deep Learning Models for Multilingual Hate Speech Detection

Deep Learning Models for Multilingual Hate Speech Detection

14 Apr 2020arXiv:2004.06465archive 2025-07-28

Sai Saketh Aluru, Binny Mathew, Punyajoy Saha, Animesh Mukherjee

Hate speech detection is a challenging problem with most of the datasets available in only one language: English. In this paper, we conduct a large scale analysis of multilingual hate speech in 9 languages from 16 different sources. We observe that in low resource setting, simple models such as LASER embedding with logistic regression performs the best, while in high resource setting BERT based models perform better. In case of zero-shot classification, languages such as Italian and Portuguese achieve good results. Our proposed framework could be used as an efficient solution for low-resource languages. These models could also act as good baselines for future multilingual hate speech detection tasks. We have made our code and experimental settings public for other researchers at https://github.com/punyajoy/DE-LIMIT.

Paper PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

punyajoy/DE-LIMIT officialmentioned in papermentioned on GitHubpytorch report

DataforGoodIsrael/DetectHateSpeech mentioned on GitHubtf report

hate-alert/DE-LIMIT mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Deep Learning Hate Speech Detection Question Similarity Zero-Shot Learning

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Hate Speech Detection	Automatic Misogynistic Identification	mBert	Accuracy	0.832	#1 of 2	Archive leaderboard	report
Question Similarity	Q2Q Arabic Benchmark	mBert	F1 score	0.8365	#3 of 3	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Adam Attention Attention Dropout BERT Dense Connections Dropout Layer Normalization Linear Layer Linear Warmup With Linear Decay Logistic Regression Multi-Head Attention Residual Connection Softmax Weight Decay WordPiece

1 archive method tag without a method page not shown.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections