Janak kapuriya
Scholar

Janak kapuriya

Google Scholar ID: -on6NR0AAAAJ
Data Science Institute, University of Galway, Ireland
Multimodal LearningVision-Language ModelsNatural Language ProcessingComputer Vision
Citations & Impact
All-time
Citations
46
 
H-index
3
 
i10-index
3
 
Publications
8
 
Co-authors
4
list available
Resume
Academic Achievements
  • Paper 'Enhancing Scientific Visual Question Answering via Vision-Caption aware Supervised Fine-Tuning' accepted at ACM Multimedia 2025 LAVA Workshop (Oct 2025).
  • Paper 'Semantic Frame Aggregation-based Transformer for Live Video Comment Generation' published in IEEE Transactions on Multimedia (Sep 2025).
  • Paper 'Exploring the Role of Diversity in Example Selection for In-Context Learning' accepted at SIGIR 2025 (Apr 2025).
  • Paper 'FlintstonesSV++: Improving Story Narration using Visual Scene Graph' accepted at ECIR Text2Story Workshop 2025 (Mar 2025).
  • Paper on Live Video Comment Generation accepted at Multimedia Transactions 2025 (Feb 2025).
  • Paper on Scientific Visual Question Answering accepted at AAAI Workshop 2025 (Dec 2024).
  • Paper 'Named Entity Recognition on Recipes' accepted at LREC-COLING 2024 (Feb 2024).
  • Paper 'MM-PhyQA: Multimodal Physics Question-Answering with Multi-image CoT Prompting' accepted at PAKDD 2024 (Jan 2024).
  • ArXiv preprint 'Spiritual-LLM: Gita Inspired Mental Health Therapy In the Era of LLMs' released in June 2025.
  • Will attend SIGIR 2025 conference in Padova, Italy, from July 12–18, 2025.
Background
  • Currently a Research Assistant at the University of Galway, Ireland, affiliated with the Insight Centre for Data Analytics.
  • Conducting research under the supervision of Prof. Paul Buitelaar in the Natural Language Processing Unit.
  • Research focuses on Factual Story Visualization.
  • Research interests span Large Vision-Language Models (LVLMs), Large Language Models (LLMs), Natural Language Processing (NLP), and Computer Vision, with an emphasis on addressing domain-specific challenges.
  • Passionate about leveraging these technologies to drive innovation in both academia and industry.