Mitigating Forgetting Between Supervised and Reinforcement Learning Yields Stronger Reasoners
Paper • 2510.04454 • Published •
1
Julianna Piskorz
jpiskorz
AI & ML interests
None yet
Recent Activity
commentedon a paper about 1 month ago
Mitigating Forgetting Between Supervised and Reinforcement Learning Yields Stronger Reasoners authored a paper 4 months ago
A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring authored a paper 4 months ago
Eliciting Numerical Predictive Distributions of LLMs Without AutoregressionOrganizations
None yet
commented a
paper about 1 month ago
commented a
paper 8 months ago