Jianguo Li

MoBE: Mixture-of-Basis-Experts for Compressing MoE-based LLMs

Understanding and Improving Length Generalization in Hierarchical Sparse Attention Models

TwinFlow: Realizing One-step Generation on Large Models with Self-adversarial Flows

Flash Multi-Head Feed-Forward Network

ACCORD: Alleviating Concept Coupling through Dependence Regularization for Text-to-Image Diffusion Personalization

WebGen-R1: Incentivizing LLMs to Generate Functional and Aesthetic Websites with Reinforcement Learning

Domain-Specific Data Synthesis for LLMs via Minimal Sufficient Representation Learning

DND: Boosting Large Language Models with Dynamic Nested Depth

Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts

From Transformer to Transponder: Introducing Contextual Modulation Training for Residual Learning in LLMs

TAROT: Test-Driven and Capability-Adaptive Curriculum Reinforcement Fine-Tuning for Code Generation

CAKE: Cascading and Adaptive KV Cache Eviction with Layer Preferences

Code Graph Model (CGM): A Graph-Integrated Large Language Model for Repository-Level Software Engineering Tasks

Rodimus*: Breaking the Accuracy-Efficiency Trade-Off with Efficient Attentions

E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning