Shijian Lu

On the Generalization Capacities of MLLMs for Spatial Intelligence

Rolling Forcing: Autoregressive Long Video Diffusion in Real Time

Skeleton-to-Image Encoding: Enabling Skeleton Representation Learning via Vision-Pretrained Models

Contrastive Self-Rewarding MLLM

Multi-Grained Policy Optimization for Multimodal Reasoning: From An Uncertainty Perspective

UniMRSeg: Unified Modality-Relax Segmentation via Hierarchical Self-Supervised Compensation

Boosting Knowledge Utilization in Multimodal Large Language Models via Adaptive Logits Fusion and Attention Reallocation

MTL-UE: Learning to Learn Nothing for Multi-Task Learning

The Curse of Multi-Modalities: Evaluating Hallucinations of Large Multimodal Models across Language, Visual, and Audio

Segment Anything with Multiple Modalities

LongHalQA: Long-Context Hallucination Evaluation for MultiModal Large Language Models

Hough Voting-based Self-Training for Vision-Language Model Adaptation