Scaling Properties of Same-Family On-Policy Distillation
Paper • 2609.32722 • Published • 326
Tamati Walker
tamati-w
·
AI & ML interests
None yet
Recent Activity
upvoted a paper 10 days ago
Scaling Properties of Same-Family On-Policy Distillation upvoted a paper 10 days ago
Selecting The Most Informative Tokens in Natural Language AutoencodersOrganizations
None yet
upvoted 4
papers 10 days ago
When Does Dense Retrieval Need Asymmetric Geometry? A Bias-Variance Theory of Shared and Dual Projections
Paper • 2609.32488 • Published • 27
Selecting The Most Informative Tokens in Natural Language Autoencoders
Paper • 2609.37040 • Published • 18
ROSS: Relearning from Self-Generated Rollouts through Selective Supervision
Paper • 2609.35954 • Published • 53