Xiaobo Xia

NExT-OMNI: Towards Any-to-Any Omnimodal Foundation Models with Discrete Flow Matching

Towards Robust Graph Unlearning via Gradient Consistency Control

Data Selection for LLM Reinforcement Learning with Improved Gradient Alignment

From One to Many: Trajectory Invariant Learning for Multimodal Large Language Model Editing

AUHead: Realistic Emotional Talking Head Generation via Action Units Control

Benchmarking and Advancing Quantization-Aware Training for Reasoning Models

Do All Individual Layers Help? An Empirical Study of Task-Interfering Layers in Vision-Language Models

GUI-R1: A Generalist R1-Style Vision-Language Action Model For GUI Agents

OFFSIDE: Benchmarking Unlearning Misinformation in Multimodal Large Language Models

VCM: Vision Concept Modeling with Adaptive Vision Token Compression via Instruction Fine-Tuning

L-MTP: Leap Multi-Token Prediction Beyond Adjacent Context for Large Language Models

DEEM: Diffusion models serve as the eyes of large language models for image perception

Continual Multimodal Contrastive Learning

UtilGen: Utility-Centric Generative Data Augmentation with Dual-Level Task Adaptation

OpenOmni: Advancing Open-Source Omnimodal Large Language Models with Progressive Multimodal Alignment and Real-time Emotional Speech Synthesis

DreamDPO: Aligning Text-to-3D Generation with Human Preferences via Direct Preference Optimization

MMEvol: Empowering Multimodal Large Language Models with Evol-Instruct