Han Liu

15

Papers

74

Total Citations

Papers (15)

STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction

Fundamental Limits of Prompt Tuning Transformers: Universality, Capacity and Efficiency

Attention Mechanism, Max-Affine Partition, and Universal Approximation

NeurIPS 2025arXiv

Liberating Seen Classes: Boosting Few-Shot and Zero-Shot Text Classification via Anchor

Fast and Low-Cost Genomic Foundation Models via Outlier Removal

Multi-Label Few-Shot Image Classification via Pairwise Feature Augmentation and Flexible Prompt Learning

On Computational Limits of Modern Hopfield Models: A Fine-Grained Complexity Analysis

Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models

Improving Accuracy and Calibration via Differentiated Deep Mutual Learning

Outlier-Efficient Hopfield Layers for Large Transformer-Based Models

EGSRAL:An Enhanced 3D Gaussian Splatting Based Renderer with Automated Labeling for Large-Scale Driving Scene

Uncertainty-Aware Contrastive Learning with Hard Negative Sampling for Code Search Tasks

A Goal Interaction Graph Planning Framework for Conversational Recommendation

VQAttack: Transferable Adversarial Attacks on Visual Question Answering via Pre-trained Models

BiSHop: Bi-Directional Cellular Learning for Tabular Data with Generalized Sparse Modern Hopfield Model