About Me

👋 About Me

I’m Arda Yüksel, a PhD Candidate at TU Darmstadt, affiliated with UKP Lab and TrustHLT. My research focuses on Multimodal Fact-Checking and Large Language Models (LLMs). I bring over three years of experience in Natural Language Processing (NLP), Applied AI, and Safe Machine Learning, with a strong focus on building robust, interpretable, and practical AI systems.

Previously, I worked as an AI Developer at Neuland AI, where I designed Retrieval-Augmented Generation (RAG) based ChatBots using Vector Databases and Knowledge Graphs. At Fraunhofer IKS, I contributed to the design and testing of safe ML systems for trustworthy AI applications. My journey into AI began at Bilkent University as an undergraduate researcher in KOÇLAB, where I worked on word embeddings, fine-tuning transformer models, and LLM benchmarking.

🧪 Research Interests

• Multimodal NLP & Fact Verification
• Large Language Models (LLMs) & Evaluation Benchmarks
• Safe & Trustworthy AI
• Retrieval-Augmented Generation (RAG)
• Knowledge Graphs & Vector Databases

📝 Publications

A full list of my publications is available on Google Scholar. My work covers NLP evaluation, applied ML systems, and trustworthy AI research.

🛠️ Tools & Technologies

Languages: Python, Java, C++, Matlab, TeX, Shell
Libraries/Frameworks: PyTorch, HuggingFace, FastAPI, Docker, Git, LlamaIndex, Phoenix
Cloud & DevOps: Azure, Docker, SSH, SLURM

📁 Datasets

• TurkishMMLU – A Turkish-language benchmark for evaluating LLMs

📫 Contact

I’m always open to discussing collaborations or sharing thoughts on NLP, machine learning, and research.
Feel free to reach out via LinkedIn.