Debolena Basak
Ph.D. Student
Artificial Intelligence
IIT Hyderabad
debolena07@gmail.com
ai20resch11003@iith.ac.in
Embedded Files
About Me
Hi! I am a final-year research scholar in the Department of Artificial Intelligence at the Indian Institute of Technology Hyderabad (IITH). As a part of the Natural Language and Information Processing (NLIP) Lab, I conduct my research under the supervision of Dr Maunendra Sankar Desarkar and Dr Srijith P.K. Recently, I completed a Research PhD Internship on Multimodal Document Understanding at Adobe Research Bangalore. My current research focuses on grounded multimodal reasoning in vision-language models and their interpretability.
Research Interests
Multimodal Document Understanding
Mechanistic Interpretability
Large Language Models (LLMs)
Large Vision Language Models (LVLMs)
Image Captioning
Retrieval Augmented Generation (RAG)
News
Glad to share that my research internship work at Adobe Research, "Diagnosing Evidence Utilization in Multimodal Document Question Answering", has been accepted at ACM SIGKDD 2026 Research Track !!
Successfully completed the Research Internship at Adobe Research, Bangalore.
Excited to join as a Research PhD Intern at Adobe Research, Bangalore! [June 2, 2025]
Honoured with the Research Excellence Award at IIT Hyderabad on 17th Foundation Day! [April 2, 2025]
Presented our paper at WACV 2025 in Tucson, Arizona, USA. Had a wonderful experience at the conference and United States! [March 2, 2025]
Happy to share that I have been selected for the WACV 2025 travel award, consisting of paper registration fee waiver and a travel grant of 1500 USD!
Happy to share that our paper - "Aerial Mirage: Unmasking Hallucinations in Large Vision Language Models", has been accepted to WACV 2025 (round 1, acceptance rate 12.1%).
Delivered an Oral Presentation of our paper at PAKDD 2024 in Taipei, Taiwan. Had an amazing experience at the conference and Taiwan!
Poster presentation at TiHAN, IIT Hyderabad.
Pleased to share that our paper titled "Transformer based Multitask Learning for Image Captioning and Object Detection" has been accepted for Oral presentation at PAKDD 2024!
Participated in IndoML 2023 at IIT Bombay. It was a wonderful experience to meet talented Indian researchers from all around the globe!
Served as a student volunteer for the prestigious ACML 2022 conference.
Publications
Diagnosing Evidence Utilization in Multimodal Document Question Answering
Debolena Basak, Digbalay Bose, Koustava Goswami, Maunendra Sankar Desarkar
32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (SIGKDD) 2026.
Aerial Mirage: Unmasking Hallucinations in Large Vision Language Models
Debolena Basak, Soham Bhatt, Sahith Kanduri, and Maunendra Sankar Desarkar
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025.
Transformer based Multitask Learning for Image Captioning and Object Detection
Debolena Basak, P.K. Srijith, and Maunendra Sankar Desarkar
Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2024)
Page updated
Report abuse