Distilling 100B+ Models 40x Faster with TRL
📝
95
TRL distillation for 100B+ teachers, 40x faster
Hang Wu
Leo-WU
AI & ML interests
None yet
Recent Activity
upvoted a paper about 21 hours ago
WorldMemBench: A Comprehensive Benchmark for Memory of Video World Models upvoted a paper about 2 months ago
TTPO: Test-Time Policy Optimization upvoted a paper 6 months ago
SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric EnvironmentsOrganizations
None yet
liked a
Space 6 months ago
liked 2
datasets over 1 year ago