5 papers
On the Nonlinearity of Learning Rate Scaling for LLM Training
Zaiwen Yang, Huaqing Zhang, Jing Xu +1
Learning-rate transfer can reduce the cost of training large language models: instead of sweeping learning rates at target scale, practitioners extrapolate from smaller runs. Exist…
Understanding Nonlinear Implicit Bias via Region Counts in Input Space
Jingwei Li, Jing Xu, Zifan Wang +2
One explanation for the strong generalization ability of neural networks is implicit bias. Yet, the definition and mechanism of implicit bias in non-linear contexts remains little…
NeuralDB: Scaling Knowledge Editing in LLMs to 100,000 Facts with Neural KV Database
Weizhi Fei, Hao Shi, Jing Xu +7
Efficiently editing knowledge stored in large language models (LLMs) enables model updates without large-scale training. One possible solution is Locate-and-Edit (L\&E), allowing s…
Scalable Model Merging with Progressive Layer-wise Distillation
Jing Xu, Jiazheng Li, Jingzhao Zhang
Model merging offers an effective way to integrate the capabilities of multiple fine-tuned models. However, the performance degradation of the merged model remains a challenge, par…
Functionally Constrained Algorithm Solves Convex Simple Bilevel Problems
Huaqing Zhang, Lesi Chen, Jing Xu +1
This paper studies simple bilevel problems, where a convex upper-level function is minimized over the optimal solutions of a convex lower-level problem. We first show the fundament…