Zhen Li
4 papers · Latest:
Computer Vision
D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models
D-OPSD introduces an on-policy self-distillation method for continuously fine-tuning efficient few-step diffusion models without losing their speed.
2605.05204
Multi-fidelity surrogates for mechanics of composites: from co-kriging to multi-fidelity neural networks
This review surveys multi-fidelity surrogate models, from co-Kriging to neural networks, for efficient and accurate prediction in composite mechanics.
2605.02871
Computer VisionUnified Map Prior Encoder for Mapping and Planning
UMPE is a Unified Map Prior Encoder that effectively fuses diverse map priors with BEV features for improved autonomous driving mapping and planning.
2605.02762
Machine LearningA Direct Approach for Handling Contextual Bandits with Latent State Dynamics
This paper introduces a direct approach for contextual bandits with latent state dynamics, achieving stronger high-probability regret bounds.
2604.08149
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.