Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
August 17, 2025

Leveraging Multilingual BERT for Contextually Coherent Dialogue Systems in Low-Resource Settings

View Full Paper
Ask AI
Bookmark
Share

Authors

HAHamza AsifRMRashid MahmoodMIMuhammad Waseem Iqbal

Discussion

Loading...

Member takes

Overview

Experimental analysis reveals improved dialogue generation results in Urdu, suggesting mBERT's effectiveness in low-resource scenarios.

Key Points

  • The mBERT-based dialogue generation system achieved a BLEU score of 0.63, outperforming the LSTM baseline by 5%.
  • Coherence scores also improved with mBERT, showing a 7% increase to 0.82 in comparison to the baseline model.
  • A novel parallel corpus of 12,000 Urdu–English conversational pairs was curated for evaluation and training purposes.
  • Findings highlight the potential of multilingual transformers for developing high-fidelity dialogue systems in underrepresented languages.

Cite This Study

Asif et al. (2025) studied this question.

synapsesocial.com/papers/68a36ddf0a429f79733313dchttps://doi.org/10.63075/yzm4vj28
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Multilingual Generative AI Framework for Urdu and Regional Language Understanding Using Large Language Models2026
  2. 2LLM-Generated Synthetic Data for Low-Resource Urdu–English Neural Machine Translation2026
  3. 3Benchmarking Pre-trained Large Language Models' Potential Across Urdu NLP tasks2024
  4. 4Multimodal Emotion Detection in Low-Resource Languages Using Lightweight Transformer Architectures: A Dual-Level Fusion Framework Integrating DistilBERT, CNN-BiGRU, and MobileViT for Efficient Real-Time Urdu Affective Computing2026
  5. 5A transformer-based Urdu image caption generation2024 · 2 citations