Maximilian Schall

Maximilian Schall

PhD Student

Hasso Plattner Institute

Biography

My research focuses on adapting language models in diverse ways. My background in software engineering is instrumental in my research approach, as it allows me to understand the complex details of language models and engineering required, while contributing innovative solutions to the field. I am a Ph.D. student at the chair of Artificial Intelligence and Intelligent Systems, where I am supervised by Gerard de Melo.

My primary work involves experimenting with and refining language models to ensure their flexibility and effectiveness across various contexts and languages. This includes developing methodologies to adapt these models to better understand and generate natural language, as well as researching parameter-efficient fine-tuning strategies to enhance their performance.

In addition to my research, I’m involved in a project that uses deep learning to track and protect gorillas in the Republic of Congo. This project merges my technical expertise with a passion for wildlife conservation. It’s a personal mission of mine to show how technology can serve as a powerful tool in the hands of those committed to protecting our planet’s natural heritage.

Publications

(2024). CommitBench: A Benchmark for Commit Message Generation. In SANER 2024.

PDF Code Dataset

(2024). NextLevelBERT: Investigating Masked Language Modeling with Higher-Level Representations for Long Documents. On ArXiv.

PDF

(2023). Efficient Parallelization Layouts for Large-Scale Distributed Model Training. WANT.

PDF

(2023). GorillaVision- Open-Set Re-Identification of Wild Gorillas. In Third International Workshop on Camera Traps, Artificial Intelligence, and Ecology, Jena, Germany.

PDF Code