2 papers
cs.AI2026
UniScale: Adaptive Unified Inference Scaling via Online Joint Optimization of Model Routing and Test-Time Scaling
Kaiyu Huang, Xingyu Wang, Mingze Kong +6
In real-world deployments of large language models (LLMs), balancing inference quality and computational cost has become a central challenge. Existing approaches tackle this trade-…
cs.CV2025
ZO-DARTS++: An Efficient and Size-Variable Zeroth-Order Neural Architecture Search Algorithm
Lunchen Xie, Eugenio Lomurno, Matteo Gambella +4
Differentiable Neural Architecture Search (NAS) provides a promising avenue for automating the complex design of deep learning (DL) models. However, current differentiable NAS meth…