LiteStage: Latency-aware Layer Skipping for Multi-stage Reasoning
Paper • 2510.14211 • Published • 9 •
2
Jiwon Song
jiwonsong
AI & ML interests
Efficient AI | Ph.D Student @ SNU-VLSI
Recent Activity
upvoted a paper about 6 hours ago
WaveFront Decoding: Parallelized Self-Speculative Decoding for Looped Language Models upvoted a paper about 6 hours ago
KVCMAS: Efficient KV cache Correction for Shared Context in Multi-Agent SystemsOrganizations
commented a
paper 12 months ago
What is the best practice of the sampling parameters for this model?
#2 opened over 1 year ago
by jiwonsong
commented a
paper over 1 year ago