Xiaojie Wang

Reading Images Like Texts: Sequential Image Understanding in Vision-Language Models

ChainGPT: Dual-Reasoning Model with Recurrent Depth and Multi-Rank State Updates

An Information-Theoretic Parameter-Free Bayesian Framework for Probing Labeled Dependency Trees from Attention Score

Fine-Tuning is Subgraph Search: A New Lens on Learning Dynamics

Probability-based Reward Value Combination Method for Multi-Objective Alignment

SMAR: Soft Modality-Aware Routing Strategy for MoE-based Multimodal Large Language Models Preserving Language Capabilities

TAMP: Task-aware Multimodal Pre-Interaction for fine-grained Large Language Models

Partial-Correlation Learning for Large Language Models with Skip-Tuning

ChunkLLM: A Lightweight Pluggable Framework for Accelerating LLMs Inference

Steering Merged LLMs for Multilingual Reasoning with Coefficient Optimization

Learning from Examples and Self-Exploration: A New Paradigm for Dynamic Fusion

Test-time Zero-shot Recognition with Good Attributes