|
|
Abhay Bhandarkar I'm currently a Business Guest at Microsoft Research India working on Computer Vision in applications for design and creativity and am fortunate to be supervised by Dr. Vineeth N Balasubramanian. I am pursuing my Bachelor's in Computer Science at Ramaiah Institute of Technology in Bangalore, India, and I work on applied computer vision, multimodal AI, and explainability in deep learning systems. My research interests span applied computer vision, perception and interpretation, explainability, and multimodality. I'm particularly fascinated by how we can make AI systems more interpretable and trustworthy, especially in high-stakes domains like autonomous systems, medical imaging and aerospace. I've worked on projects ranging from medical image segmentation with efficient vision transformers to defense surveillance systems using deep learning, and I'm passionate about bridging the gap between cutting-edge research and real-world applications. |
|
|
||
सुखार्थिनः कुतो विद्या, नास्ति विद्यार्थिनः सुखम्। सुखार्थी वा त्यजेद्विद्यां, विद्यार्थी वा त्यजेत्सुखम्|| "The seeker of comfort has no knowledge, and the seeker of knowledge has no comfort. One who desires comfort must abandon knowledge, and one who desires knowledge must abandon comfort." ||
|
🎯 Recent Updates |
🏆 Awards & Achievements
📅 Timeline
|
ResearchI'm interested in computer vision, deep learning, multimodal AI, and explainability. My work focuses on developing efficient and interpretable models for real-world applications. Some papers are highlighted. |
Publications |
|
|
Causal Quantification of the Sensitivity-Reliability Trade-Off in Semantic XAI: Comparing Object-Aware (SAM) and Texture-Aware (SLIC) Segmentation
Abhay Bhandarkar, et al. AAAI AI4Science Workshop, 2026 (Accepted, To Appear) Causal quantification framework comparing object-aware and texture-aware segmentation approaches for understanding the sensitivity-reliability trade-off in semantic explainable AI. |
|
|
SIMHA: Sustainable Integrated Microbial and Hydrogen Fuel Cell Architecture
SA Kumar, Abhay Bhandarkar, Dr. Nagaraju Kottam, et al. IEEE SPACE, 2025 A novel sustainable energy architecture integrating microbial fuel cells with hydrogen fuel cells for space and terrestrial applications. |
|
|
Augmenting URL-BERT based Phishing Detection with Favicon Metadata
Abhay Bhandarkar, Vishwachetan D, Geetha B ICCTRDA, 2025 (Accepted, To Appear) A novel approach augmenting URL-BERT based phishing detection with favicon metadata for improved accuracy in identifying malicious websites. |
|
|
Unmasking the AI Hand: A Machine Learning Approach to Deciphering Authorship
Abhay Bhandarkar, Vishwachetan D IEEE INOCON, 2024 A machine learning framework for detecting AI-generated text and distinguishing between human and machine authorship in written content. |
|
|
Advancements in Defense Surveillance: Integrating Deep Learning and Enhanced Materials in CubeSAT Systems
Abhay Bhandarkar, Riddhi Rai, Sushma Bylaiah IEEE SPACE, 2024 Integrating state-of-the-art deep learning architectures with advanced materials science for next-generation CubeSAT-based defense surveillance systems. |
|
|
TotCareBot: A Comprehensive Approach towards Enhancing Early Childhood Education through Robotics
Manjula R Chougala, Sushma Bylaiah, Abhay Bhandarkar, Akshata S Bhayyar, P N Pavitra I4C, 2024 A comprehensive robotic system designed to enhance early childhood education through interactive and engaging learning experiences. |
|
|
Minds and Materials in Flight: The Dynamic Duo of Smart Materials and Machine Learning in Aerospace Evolution
Abhay Bhandarkar, Dr. Nagaraju Kottam, Vishwachetan D ICAMES, 2023 Exploring the synergistic integration of smart materials and machine learning for advancing aerospace engineering and design. |
Pre-prints & In Writing |
|
|
Topic Modelling with BERTopic: Understanding the likings of humans across various LLMs via thorough analysis of LMSYS dataset
Abhay Bhandarkar, Khushi, Gaurav Mishra, et al. arXiv arXiv / PDF Applies BERTopic to the LMSYS-Chat-1M dataset to uncover thematic patterns in how users interact with and evaluate competing large language models. |
|
|
MITRA: Model In The Rear Approach to Safeguarding LLMs for low cost and scalable solutions
Abhay Bhandarkar, et al. In Writing A novel approach for safeguarding LLMs with low-cost and scalable guardrail solutions. |
Patents |
|
📜
|
JAM Algorithm for Probabilistic String Search
Abhay Bhandarkar Patent Application Submitted Application No.202541017132 A A novel algorithm for efficient probabilistic string search with applications in data retrieval. |
|
📜
|
SIMHA: Sustainable Integrated Microbial and Hydrogen Fuel Cell Architecture
Abhay Bhandarkar Patent Application Submitted Application No.202541017133 A A sustainable energy solution integrating microbial and hydrogen fuel cell technologies. |
|
© 2025 Abhay Bhandarkar | Last updated: November 2025
|