← Search

Aoxuan Silvia Zhang

2 accepted papers

2026

LS-Merge: Merging Language Models in Latent Space

ICLR 2026poster

Model merging in weight space is an efficient way to reuse pretrained models, but existing methods typically assume matching architectures or sizes, making heterogeneous merges brittle or infeasible. We address this limitation by encoding model weights into a smooth latent space, enabling cross-arch…

Cited by 0SourcecodeScholar
2025

Cost-Sensitive Freeze-thaw Bayesian Optimization for Efficient Hyperparameter Tuning

NeurIPS 2025poster

In this paper, we address the problem of cost-sensitive hyperparameter optimization (HPO) built upon freeze-thaw Bayesian optimization (BO). Specifically, we assume a scenario where users want to early-stop the HPO process when the expected performance improvement is not satisfactory with respect to…

Cited by 0SourceScholar