November 2025 🚨 Paper Alert: Our paper "Chain-of-Thought Driven Adversarial Scenario Extrapolation for Robust Language
Models" got accepted at AAAI 2026
October 2025 🎉 Our paper "From Insight to Exploit: Leveraging LLM Collaboration for Adaptive Adversarial Text
Generation" is published in EMNLP 2025
October 2025 🚨 Paper Alert: Our paper "Attention Pruning: Automated Fairness Repair of Language Models via Surrogate
Simulated Annealing" got accepted at ICSE 2026
September 2025 🎯 Completed my summer internship for the second time at Mitsubishi Electric Research Laboratory
August 2025 🚨 Paper Alert: Our paper "From Insight to Exploit: Leveraging LLM Collaboration for Adaptive Adversarial
Text Generation" got accepted at EMNLP 2025
August 2025 ✍🏽 Invited as a Program Committee (PC) member at AAAI 2026
July 2025 🎉 Reached 200 citations for my research works at Google Scholar
June 2025 ✍🏽 Invited as a Program Committee (PC) member at AAAI/ ACM AIES 2025
Secure and Robust Machine Learning
With the widespread deployment of Machine learning (ML) algorithms, currently, their security and transparency concerns demand more attention than ever before. As the primary field of my Ph.D. research, I investigate the vulnerabilities of state-of-the-art ML models that might invoke malicious intrusion to compromise their fidelity and security.
Natural Language Processing
Some of my recent works highly focused on different natural language processing tasks for both zero-shot and few-shot learning, such as language modeling (text generation and translation), reasoning, Instruction tuning and Alignment. I am continuously learning cutting-edge techniques to implement well-performant models for NLP applications.
Federated Learning
I work with a decentralized implementation of ML called federated learning, where edge devices can attend ML training on purpose by ensuring data privacy. Besides having a sufficient theoretical background, I am also flexible in building custom FL models for different applications in NLP and computer vision. I use Pytorch and Tensorflow-Federated libraries for implementing and fine-tuning the FL configurations.
Applied Machine Learning
Besides having a thorough understanding of the theoretical aspects of Machine Learning algorithms, I extensively apply them in several domains, e.g., Diffusion & multimodal learning, Computer Vision, and Audio signal Processing. I am also flexible in fine-tuning model architectures for both general and ad-hoc transfer learning applications. For this purpose, I gathered hands-on experience in using different machine learning and deep learning libraries in Python, such as Pytorch, Tensorflow, Transformers and Scikit-learn.
Python Programming
I carry 6+ years of experience in Python programming with an advanced level of expertise. I am also mentoring in an undergraduate-level Python course as a part of my graduate assistantship at Penn State, CSE. Besides using Python for problem-solving, I extensively use it for my research work in machine learning, security, and data science.
