QED-Nano: Teaching a Tiny Model to Prove Hard Theorems
📝
56
Who needs 1T parameters? Olympiad proofs with a 4B model
Justin Wang
Justin6657
AI & ML interests
None yet
Recent Activity
liked
a Space 11 days ago
lm-provers/qed-nano-blogpost upvoted a paper 26 days ago
Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning updated
a model 9 months ago
Justin6657/PoPilot Organizations
liked a Space 11 days ago