Hello there, I’m Aswini Kumar 🙂

I am a computer science researcher passionate about developing robust large language models (LLMs) and exploring the exciting potential of Diffusion Language Models. My curiosity drives me to push the boundaries of what AI can safely and reliably achieve.

What am I working on now?

Currently, I am a doctoral fellow under the guidance of Prof. Tanmoy Chakraborty at Laboratory for Computational Social Systems (LCS2), Indian Institute of Technology, Delhi. My PhD thesis asks a crucial question: How can we counter harmful prompts and prevent language models from generating unsafe or damaging responses? My research delves deep into the vulnerabilities of natural language processing systems, seeking innovative solutions to make conversational AI more trustworthy and secure.

Where have I come from?

My academic journey began at Veer Surendra Sai University of Technology, Burla, where I earned my bachelor’s degree. I then pursued my master’s at ABV Indian Institute of Information Technology, Gwalior, where I was honored with the Best Postgraduate Merit Award by the Hon’ble President of India, Smt. Droupadi Murmu. Each step has fueled my commitment to impactful research and academic excellence.

What’s next?

As I continue my doctoral research, I am eager to collaborate, learn, and contribute to the evolving landscape of AI safety and language technologies.