Location: Saarbrücken

I graduated from Sharif University of Technology in Computer Engineering in 2024, and I’m now pursuing my MSc in Computer Science (DSAI) at Saarland University. My current focus is LLM theory and AI safety, building on a background in Natural Language Processing, Machine Learning, and Human-Centered Computing.

Much of my published work sits in human-centered computing and NLP, from multimodal retrieval and accessibility to how people experience conversational systems. More recently, I’ve been moving toward understanding the foundations of large language models and how to make them safer and more reliable. Alongside research, I spent three years as a software engineer building products at scale.

Hi! I am Hurriya, a Master’s student in Computer Science at Saarland University. My interests lie in Natural Language Processing, LLMs, and the intersection of data and society. I am especially interested in how these systems can solve real-world problems while remaining transparent and accessible.
Alongside my studies, I work as a student research assistant (HiWi) at Saarland University and as a working student in data engineering. Before my Master’s, I worked as a software engineer for 2 years.
During my studies, I have worked on projects including an AI-powered hospital cost analysis assistant and the detection of LLM-generated content, while also exploring RAG systems and agent-based approaches.
In the long run, I aim to work at the intersection of AI research and entrepreneurship, building systems that create meaningful impact.
I am a Computer Science Master’s Student at Saarland University and a Research Assistant at the Department of Language Science and Technology in Saarbrücken. My academic work centers on Large Language Models (LLMs) and Trustworthy AI. Specifically, I focus on developing methods to steer model behavior without the need for heavy retraining and implementing constrained generation techniques with IPA.

Before my current academic roles, I built a strong foundation in software and data engineering in Nepal. I hold a Bachelor’s degree in Computer Engineering from Kathmandu University.

I am working as a research assistant/HiWi at the Max Planck Institute for Informatics in the D2 group, led by Prof. Dr. Bernt Schiele where I work with my colleagues on vision-language models, study their behaviour on multi-modal documents and work on improving them. Apart from vision-language models, I am also interested in explainbale machine learning. I also worked on a small research project that analyzed time-steps of a diffusion models to test the appearance of concepts across time-steps.

Previously, I had the opportunity to do a research internship at Bennett University (India), where my supervisor and I developed a light-weight image encryption algorithm for IoT applications. The paper is titled “IETD: a novel image encryption technique using Tinkerbell map and Duffing map for IoT applications”, published in Springer Nature, Multimedia tools and applications.

Before pursuing my master’s studies, I worked at Deloitte US-India for more than two years. There, my work focused on automating enterprise operations and research solutions for new use-cases apart from front-end development.

Keziah Reina is an M.Sc. student in Language Science and Technology at Saarland University and a Research Assistant at the Societal Observatory Using Novel Data Sources (SOUNDS), led by Dr. Ingmar Weber. Previously, she served as an AI Safety Saarland Research Fellow under Dr. Siddharth Mehrotra, designing experiments on AI trust calibration using insurance-inspired mechanisms. Her research interests sit at the intersection of multilingual NLP, ethical AI, and tech for underserved communities. Prior to her graduate studies, she worked as a Linguistic Data Specialist at QuillBot AI, developing Spanish-language NLP tools. She holds a B.A. in Linguistics with High Honors from UT Austin, where she worked as a Computational Linguistics Research Assistant. She has been recognized on the Dean’s Honor List, Presidential Academic Honor List, and as a College Honors Scholar at UT Austin, as well as an EMNLP 2023-accepted co-author on Multilingual Simplification of Medical Texts.
I am a Computer Science graduate and current Master’s student in Data Science and Artificial Intelligence at Saarland University with hands-on experience developing large-scale defense simulation and interoperability systems at National Aerospace Technology Park. Skilled in C++, C#, Python, SQL, and system-level software engineering, with strong foundations in Data Structures, Object-Oriented Programming, distributed systems, and high-performance application development. My work on industrial-scale real-time simulation systems strengthened my interest in scalable machine learning systems, simulation-based inference, and transdisciplinary AI applications. I also developed a Multilingual Speech Emotion Recognition system, focusing on cross-lingual emotion classification using CNNs and MLPs. My research interests include probabilistic inference, multimodal NLP, simulation-based learning, and data-driven intelligent systems that bridge robust software engineering with advanced AI research.
I am a Computer Science and NLP student with a strong research interest in the interpretability and reliability of large language models. My work focuses on mechanistic interpretability, sparse autoencoders, and the analysis of internal model representations. In several research projects, I investigated negation-related circuits in Phi-2 using sparse autoencoders and currently study copy-based in-context learning behavior in GPT-2 and Gemma-2.

My Bachelor’s thesis explored framing generation through in-context learning, examining how prompt variations influence model behavior. I also conducted a reproduction study on non-conservative force models in atomic simulations, strengthening my experience in experimental evaluation and reproducible research. Technically, I work primarily with Python, PyTorch, TransformerLens, and LogitLens. My broader goal is to contribute to trustworthy and interpretable AI systems by improving our understanding of how language models process and represent information.

I am currently pursuing a Masters degree in Data Science and Artificial Intelligence from Universität des Saarlandes.

I have over 2.5 years of professional experience as a Machine Learning Engineer at Arbisoft, a leading software company in Pakistan. I have worked on solving various problems including NLP, time series forecasting and math or stats based models. I have hands on experience of solving real-world problems along with understanding, cleaning and using large amounts of data in various AI applications using GCP services and data warehouses like Snowflake and BigQuery.

Prior to this, I completed my bachelors degree in software engineering from COMSATS University Islamabad. In my final year project, I developed a mobile application using concepts of NLP. machine learning, deep learning and web scraping. I graduated with a silver medal, showcasing my excellent academic record.

GitHub profile for personal projects: https://github.com/fatimahanif

I am a Professor of Computational Linguistics in the Department of Language Science and Technology at Saarland University. I am also a Professor by courtesy of Computer Science and Associate Faculty at the Max Planck Institute for Informatics.

I lead the Computational Linguistics group and am the speaker of the Research Training Group “Neuroexplicit Models of Language, Vision, and Action”. I am an ACL Fellow.

My PhD is in Computer Science from Saarland University; my advisors were Manfred Pinkal and Gert Smolka. Before coming back to Saarbrücken, I was a postdoc at Columbia University and the University of Edinburgh and a Professor of Theoretical Computational Linguistics at the University of Potsdam. I have spent sabbaticals at Meta AI in Paris and the Allen Institute for Artificial Intelligence (AI2) in Seattle.

I am a huge fan of Typst and try to do as much of my writing as possible in Typst rather than LaTeX. I maintain the tracl template for *ACL papers and the Pergamon bibliography package. I also maintain the Typstit tool for dragging Typst formulas into other apps. In my spare time, I am a certified coach for a cappella groups and was an internationally certified contest judge in the Musicality category from 2023-2026.

My research is at the intersection of Machine Learning, Natural Language Processing, and Neuroscience, with a focus on building computational models of language processing in the brain that can also improve natural language processing systems.