Xuming Hu

10

Papers

103

Total Citations

1

Affiliations

Affiliations

HKUST

Papers (10)

Dataset Distillation with Neural Characteristic Function: A Minmax Perspective

OmniSAM: Omnidirectional Segment Anything Model for UDA in Panoramic Semantic Segmentation

Towards Understanding Factual Knowledge of Large Language Models

ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference

RealRAG: Retrieval-augmented Realistic Image Generation via Self-reflective Contrastive Learning

Can Watermarked LLMs be Identified by Users via Crafted Prompts?

ICT: Image-Object Cross-Level Trusted Intervention for Mitigating Object Hallucination in Large Vision-Language Models

Three Heads Are Better than One: Improving Cross-Domain NER with Progressive Decomposed Network

Reducing Unimodal Bias in Multi-Modal Semantic Segmentation with Multi-Scale Functional Entropy Regularization

LazyMAR: Accelerating Masked Autoregressive Models via Feature Caching