Wenwu Zhu

SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Acceleration

EvoAgent: Self-evolving Agent with Continual World Model for Long-Horizon Tasks

Adaptive Mixture of Disentangled Experts for Dynamic Graph Out-of-Distribution Generalization

WiSparse: Boosting LLM Inference Efficiency with Weight-Aware Mixed Activation Sparsity

Reasoning Diffusion for Unpaired Text-Image to Video Generation

Cross-granularity Video LLM for Long Video Understanding

A Unified Understanding and Generation Framework for Ego-Centric Tracing in Dynamic World

Embodied Graph: A Relational World Model

AutoGFM: Automated Graph Foundation Model with Adaptive Architecture Customization

Dynamic Mixture of Curriculum LoRA Experts for Continual Multimodal Instruction Tuning

Out-of-Distribution Generalized Graph Anomaly Detection with Homophily-aware Environment Mixup

Disentangling Invariant Subgraph via Variance Contrastive Estimation under Distribution Shifts

Self-supervised Masked Graph Autoencoder via Structure-aware Curriculum

Modularized Self-Reflected Video Reasoner for Multimodal LLM with Application to Video Question Answering

Causal-aware Graph Neural Architecture Search under Distribution Shifts

MSR-ViR: Modularized Self-reflected Video Reasoner for Video Question Answering

Self-supervised Masked Graph Autoencoder via Structure-aware Curriculum

Evaluating the Generalization Ability of Quantized LLMs: Benchmark, Analysis, and Toolbox

MindDETR: Beyond Semantics, Exploring Positional Cues from Brain Activity

Curriculum GNN-LLM Alignment for Text-Attributed Graphs

GAQAT: Gradient-Adaptive Quantization-Aware Training for Domain Generalization

Accelerate Quantization Aware Training for Diffusion Models with Difficulty-aware Time Allocation

Text-to-graph Generation with Conditional Diffusion Models Guided by Graph-aligned LLMs

CaLMol: Disentangled Causal Graph LLM for Molecular Relational Learning