Chang ZENG
Chang ZENG
Home
News
Publications
Skills
Activities
Experience
Posts
Light
Dark
Automatic
Posts
Adaptive Granularity Importance Sampling for Policy Optimization
A research note on adaptive granularity importance sampling for policy optimization, focusing on speech token dependencies and segment-level weighting strategies such as MASPO-Fixed, MASPO-LogProb, and MASPO-TokenVal.
Chang ZENG 曾畅 曾 暢 (ソウ チョウ)
Mar 11, 2026
9 min read
Research Notes
Cite
×
