Qianxiao Li

The Effect of Attention Head Count on Transformer Approximation

Machine Unlearning under Retain–Forget Entanglement

The Effect of Depth on the Expressivity of Deep Linear State-Space Models

LoRA in the Right Place: Which Block to Tune in Parameter-Efficient Fine-Tuning?

Allocation of Parameters in Transformers

A unified framework for establishing the universal approximation of transformer-type architectures

Autocorrelation Matters: Understanding the Role of Initialization Schemes for State Space Models

From Weight-Based to State-Based Fine-Tuning: Further Memory Reduction on LoRA with Parallel Control

Continuity-Preserving Convolutional Autoencoders for Learning Continuous Latent Dynamical Models from Images

BP-Modified Local Loss for Efficient Training of Deep Neural Networks

Rate of Approximation by Flows: A Case Study on the Eikonal Equation

Unifying Back-Propagation and Forward-Forward Algorithms through Model Predictive Control

L-MSA: Layer-wise Fine-tuning using the Method of Successive Approximations