... them to specialized reasoning domains as well as to address a generalization gap in LLM reasoning and their adaptation across reasoning domains. This work will blend cutting-edge experimentation – spanning RL, few-shot learning, meta-learning, etc. – with formal analysis to push the boundaries of what modern AI systems can reliably reason about. Position available starting around July 2026 (position may begin later as well). The duration of a PhD ...
Inserat ansehen