Dongfang Li

Is On-Policy Data always the Best Choice for Direct Preference Optimization-Based LM Alignment?

KaLM-Embedding-V2: Superior Training Techniques and Data Inspire A Versatile Embedding Model

LycheeDecode: Accelerating Long-Context LLM Inference via Hybrid-Head Sparse Decoding

SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-Reflection

In-Context Learning State Vector with Inner and Momentum Optimization