Chonghan Liu

Flatter Tokens are More Valuable for Speculative Draft Model Training

Extending Foundation Models to Low-Resource Languages: Vocabulary Expansion and Policy Optimization

d$^2$Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Caching

Rethinking LLM Ensembling from the Perspective of Mixture Models

Ariadne: Advancing Real-world Path-finding Capabilities of VLMs via Difficulty-aware Reinforcement Learning

AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence