The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment
Paper • 2604.06377 • Published • 5
Tu Vu
tuvllms
·
AI & ML interests
None yet
Recent Activity
submitted a paper 2 days ago
The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment authored a paper about 1 month ago
PRISM: Pushing the Frontier of Deep Think via Process Reward Model-Guided InferenceOrganizations
submitted a
paper to Daily Papers 2 days ago
authored a
paper about 1 month ago