James Kwok

Discrete Latent Features Ablate Adversarial Attack: A Robust Prompt Tuning Framework for VLMs

Latent-to-Data Cascaded Diffusion Models for Unconditional Time Series Generation

Multi-Order Wavelet Derivative Transform for Deep Time Series Forecasting

Reasoning-Aligned Perception Decoupling for Scalable Multi-modal Reasoning

SPARD: Defending Harmful Fine-Tuning Attack via Safety Projection with Relevance–Diversity Data Selection

DocPruner: A Storage-Efficient Framework for Multi-Vector Visual Document Retrieval via Adaptive Patch-Level Embedding Pruning

Degree-Conscious Spiking Graph for Cross-Domain Adaptation

Improving Batch Normalization in Federated Learning with Non-IID Features

Diffusion with Truncated Blocks: Towards Fast and High-Quality Text Generation using Truncated Block Generation

Keep Refining Your Discrete Diffusion Model: A Mixture of Absorbing and Uniform Processes

Channel Matters: Estimating Channel Influence for Multivariate Time Series

Joint Gradient Balancing for Data Ordering in Finite-Sum Multi-Objective Optimization

SPMDM: Enhancing Masked Diffusion Models through Simplifing Sampling Path

Multi-Objective One-Shot Pruning for Large Language Models

Pareto Merging: Multi-Objective Optimization for Preference-Aware Model Merging

Enhancing Pruned Models by Input Compensation

Curriculum-aware Training for Discriminating Molecular Property Prediction Models

Graph Vision Networks for Link Prediction

EMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions

Input Compensation for Pruned Models

Mixture of insighTful Experts (MoTE): The Synergy of Thought Chains and Expert Mixtures in Safety Self-Alignment

Open Your Eyes: Vision Enhances Message Passing Neural Networks in Link Prediction

Direct Alignment of Language Models via Quality-Aware Self-Refinement