The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment
Paper • 2604.06377 • Published • 7
Rishab Balasubramanian
rishabbala
·
AI & ML interests
None yet
Recent Activity
upvoted a paper 7 days ago
PRISM: Pushing the Frontier of Deep Think via Process Reward Model-Guided Inference updated a dataset over 1 year ago
allenai/math_qaOrganizations
None yet
upvoted a paper 3 days ago