Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
AMO: Operator-level Adaptive Muon Orthogonalization
Xinlin Zhuang, Panyi Ouyang, Yichen Li +7
Muon has recently emerged as a competitive alternative to AdamW for large-scale pre-training, with orthogonalization via Newton-Schulz (NS) iteration as its core operation. Standar…
cs.LG2025
CHIPS: Efficient CLIP Adaptation via Curvature-aware Hybrid Influence-based Data Selection
Xinlin Zhuang, Yichen Li, Xiwei Liu +11
Adapting CLIP to vertical domains is typically approached by novel fine-tuning strategies or by continual pre-training (CPT) on large domain-specific datasets. Yet, data itself rem…
cs.LG2024
Adaptive Inference: Theoretical Limits and Unexplored Opportunities
Soheil Hor, Ying Qian, Mert Pilanci +1
This paper introduces the first theoretical framework for quantifying the efficiency and performance gain opportunity size of adaptive inference algorithms. We provide new approxim…