arxiv:2509.24510
Patrik Wolf
patrikwolf
ยท
AI & ML interests
Test-time training, preference learning, alignment, theory
Recent Activity
upvoted a paper 10 days ago
Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines? upvoted a paper 15 days ago
Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation upvoted a paper about 1 month ago
Reinforcement Learning via Self-Distillation Organizations
