The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation
Paper • 2609.36484 • Published • 553
Fiona L. Kobayasi
fionakcej
·
AI & ML interests
Deep learning, but mostly the shallow parts
Recent Activity
upvoted a paper 11 days ago
VoxMem: Benchmarking Multimodal Memory in Large Audio Language Models upvoted a paper 11 days ago
MaLiang-Harness: A Programmable Path to Image and Video GenerationOrganizations
None yet