Search:
Match:
4 results
Research Paper#Natural Language Processing, Fake News Detection, Domain Adaptation, Urdu🔬 ResearchAnalyzed: Jan 3, 2026 19:36

Urdu Fake News Classification with Domain Adaptation

Published:Dec 28, 2025 04:42
1 min read
ArXiv

Analysis

This paper addresses the critical problem of fake news detection in a low-resource language (Urdu). It highlights the limitations of directly applying multilingual models and proposes a domain adaptation approach to improve performance. The focus on a specific language and the practical application of domain adaptation are significant contributions.
Reference

“Domain-adapted XLM-R consistently outperforms its vanilla counterpart.”

PermalinkArXiv
Paper#NLP, Hope Speech Detection, Multilingual, Low-Resource Languages, Transformers🔬 ResearchAnalyzed: Jan 3, 2026 16:22

Multilingual Hope Speech Detection Framework for Low-Resource Languages

Published:Dec 27, 2025 21:23
1 min read
ArXiv

Analysis

This paper addresses the under-representation of hope speech in NLP, particularly in low-resource languages like Urdu. It leverages pre-trained transformer models (XLM-RoBERTa, mBERT, EuroBERT, UrduBERT) to create a multilingual framework for hope speech detection. The focus on Urdu and the strong performance on the PolyHope-M 2025 benchmark, along with competitive results in other languages, demonstrates the potential of applying existing multilingual models in resource-constrained environments to foster positive online communication.
Reference

“Evaluations on the PolyHope-M 2025 benchmark demonstrate strong performance, achieving F1-scores of 95.2% for Urdu binary classification and 65.2% for Urdu multi-class classification, with similarly competitive results in Spanish, German, and English.”

PermalinkArXiv
Research Paper#AI-Generated Text Detection, Bengali Language, Transformers🔬 ResearchAnalyzed: Jan 4, 2026 00:14

Detecting AI-Generated Bengali Text: A Transformer Study

Published:Dec 25, 2025 15:04
1 min read
ArXiv

Analysis

This paper addresses the important problem of detecting AI-generated text, specifically focusing on the Bengali language, which has received less attention. The study compares zero-shot and fine-tuned transformer models, demonstrating the significant improvement achieved through fine-tuning. The findings are valuable for developing tools to combat the misuse of AI-generated content in Bengali.
Reference

“Fine-tuning significantly improves performance, with XLM-RoBERTa, mDeBERTa and MultilingualBERT achieving around 91% on both accuracy and F1-score.”

PermalinkArXiv
Research#LLM🔬 ResearchAnalyzed: Jan 10, 2026 14:26

SHROOM-CAP's Data-Centric Approach to Multilingual Hallucination Detection

Published:Nov 23, 2025 05:48
1 min read
ArXiv

Analysis

This research focuses on a critical problem in LLMs: the generation of factual inaccuracies across multiple languages. The use of XLM-RoBERTa suggests a strong emphasis on leveraging cross-lingual capabilities for effective hallucination detection.
Reference

“The study uses XLM-RoBERTa for multilingual hallucination detection.”

PermalinkArXiv