by Stefan Heinrich Papers
2 papers found
Improving Reasoning Performance in Large Language Models via Representation Engineering
Bertram Højer, Oliver Jarvis, Stefan Heinrich
ICLR 2025posterarXiv:2504.19483
15
citations
Prioritized Soft Q-Decomposition for Lexicographic Reinforcement Learning
Finn Rietz, Erik Schaffernicht, Stefan Heinrich et al.
ICLR 2024poster