Transformer-Based Ensemble Models Improve Hate Speech Detection and Sentiment Analysis in Nepali Memes
Researchers developed machine learning models to detect hate speech and analyze sentiment in Nepali-language memes, which are complicated by code-mixing and lack of established resources. The study evaluated six transformer-based models and two ensemble voting strategies on text extracted from memes using OCR. Soft Voting ensembles achieved the best results for sentiment analysis with a 15.8% improvement over baseline models, suggesting ensemble strategies should be tailored to classification task type.
A new study accepted at the 2nd Workshop on Challenges in Processing South Asian Languages addresses the underexplored problem of analyzing Nepali-language internet memes for hate speech and sentiment. The research team extracted embedded text from memes using optical character recognition (OCR) and applied transformer-based neural network architectures to classify content. The study compared six distinct models and tested both Hard and Soft Voting ensemble aggregation methods across two tasks: binary hate speech detection and three-class sentiment analysis. Results showed that a standalone decoder-only transformer model performed best for binary classification, while Soft Voting ensembles excelled at multi-class sentiment tasks, achieving a 15.8% relative improvement in Macro F1-score compared to the strongest individual baseline. The findings underscore that different ensemble strategies are suited to different classification objectives and that careful selection of aggregation methods is critical for optimal performance.
What's missing
The study's limitations regarding dataset size, language coverage beyond Nepali, generalizability to other South Asian languages, and potential biases in the training data are not discussed in the abstract provided.
What different sources said
- arXiv cs.AICenter
TeamHerald@CHIPSAL 2026: Hate Speech Detection and Sentiment Analysis of Nepali Memes using Transformer-based Architectures and Ensemble Learning
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.