Yequan Wang

Multi-objective Large Language Model Alignment with Hierarchical Experts

Skill Weaving: Efficient Self-Improvement of LLMs via Modular Skillpacks

VisualPrompter: Semantic-Aware Prompt Optimization with Visual Feedback for Text-to-Image Synthesis

FLM-Audio: Natural Monologues Improves Native Full-Duplex Chatbots via Dual Training

Combination Generalization of Capability-Specific Neurons in LLMs

EgoMem: Lifelong Memory Agent for Full-duplex Omnimodal Models

The Price of a Second Thought: On the Evaluation of Reasoning Efficiency in Large Language Models

OmniDFA: A Unified Framework for Open Set Synthesis Image Detection and Few-Shot Attribution

BDQ: Bidirectional Diagonal Quantization for LLMs

If an LLM Were a Character, Would It Know Its Own Story? Evaluating Lifelong Learning in LLMs

EmotionTalk: An Interactive Chinese Multimodal Emotion Dataset With Rich Annotations

Effective Interaction between Quantization and Low-Rank Decomposition based on LLMs

Capability Localization: Capabilities Can be Localized rather than Individual Knowledge

Reflection on Knowledge Graph for Large Language Models Reasoning

Multi-modality Expansion and Retention for LLMs through Parameter Merging and Decoupling

Few-Shot Learner Generalizes Across AI-Generated Image Detection