Official repository of the ACL 2024 paper "Guardians of the Machine Translation Meta-Evaluation: Sentinel Metrics Fall In!".

Persian text emotion recognition by fine tuning the XLM-RoBERTa Model + Bidirectional GRU layer.

Unattended Lightweight Text Classifiers with LLM Embeddings

This study presents a novel multimodal fusion technique for disaster identification in Bangla, combining text and image data using the "BanglaCalamityMMD" dataset. Employing DisasterTextNet, DisasterImageNet, and DisasterMultFusionNet, the approach addresses a key gap in Bangla disaster research.

This study introduces MultiBanFakeDetect, a novel multimodal dataset for Bangla fake news detection, combining textual and visual information. It features TextFakeNet for text analysis and MultiFusionFake for integrating multimodal data.

Tencent Pre-training framework in PyTorch & Pre-trained Model Zoo

Punctuation Restoration for Khmer language

[KDD 2024] Improving the Consistency in Cross-Lingual Cross-Modal Retrieval with 1-to-K Contrastive Learning

Multilingual Deception Detection of GPT-generated Hotel Reviews

notebooks to finetune `bert-small-amharic`, `bert-mini-amharic`, and `xlm-roberta-base` models using an Amharic text classification dataset and the transformers library

Open Source Pre-training Model Framework in PyTorch & Pre-trained Model Zoo

Проект в рамках ВКР под названием "Разработка программного модуля для анализа документов, подтверждающих индивидуальные достижения"

This project explores the foundational concepts of ML, NLP, and model optimization to develop an efficient and user-friendly healthcare solution.

🤖 A PyTorch library of curated Transformer models and their composable components

Trankit is a Light-Weight Transformer-based Python Toolkit for Multilingual Natural Language Processing

Currently running NLP project about political communication on Twitter. You can find more projects in my portfolio.

ZaBantu is a fleet of light-weight Masked Language Models for Southern Bantu Languages

ML and Natual Language Processing

Fine tuned BERT, mBERT and XLMRoBERTa for Abusive Comments Detection in Telugu, Code-Mixed Telugu and Telugu-English.