Scholar
Till Speicher
Google Scholar ID: PGLrYWgAAAAJ
PhD student, Planck Institute for Software Systems (MPI-SWS)
Citations & Impact
Citations
631
H-index
6
i10-index
4
Publications
20
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Published multiple papers including 'Towards Reliable Latent Knowledge Estimation in LLMs: Zero-Prompt Many-Shot Based Factual Knowledge Extraction', 'Understanding Memorisation in LLMs: Dynamics, Influencing Factors, and Implications', and more.
Research Experience
- AI researcher at Aleph Alpha; during his PhD, he worked on how large language models memorize information, learn formal grammars, and acquire factual knowledge, and also on fairness in algorithmic decision-making and properties of representations in deep neural networks.
Education
- PhD from Max Planck Institute for Software Systems (MPI-SWS), Germany, advised by Krishna Gummadi; Bachelor's degree in Computer Science from Saarland University, Germany.
Background
- AI researcher with interests in understanding learning modes of large language models, fairness in algorithmic decision-making, and properties of representations in deep neural networks.
Miscellany
- Enjoys triathlon as a side hobby.
Co-authors