Jing Li

Multi-objective Large Language Model Alignment with Hierarchical Experts

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

Echoes as Anchors: Probabilistic Costs and Attention Refocusing in LLM Reasoning

Language Bias in LVLMs: From In-Depth Analysis to Simple and Effective Mitigation

Improving Code Translation Correctness and Efficiency with Multi-Perspective Exploration and Difference-Aware Selection

Knowledge Fusion of Large Language Models via Modular SkillPacks

EgoMem: Lifelong Memory Agent for Full-duplex Omnimodal Models

Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward

Vocabulary Fixation Reveals Visual Attention Sink for Hallucination Mitigation in LVLMs

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

From Vulnerability to Defense: Understanding and Mitigating MASK-Based Attacks in dLLMs

Controllable and Interpretable Multi-Value Alignment For Large Language Model

Introspective Adversarial Learning: Autonomous and Continual Preference Learning for LLM Alignment

Thinking in Character: Advancing Role-Playing Agents with Role-Aware Reasoning

Function-to-Style Guidance of LLMs for Code Translation

Learning to Watermark: A Selective Watermarking Framework for Large Language Models via Multi-Objective Optimization

NeurIPT: Foundation Model for Neural Interfaces

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