Yufei Huang

Ensuring LLMs behave in ways that reflect human ethics and values.

Examining grokking, double descent, and emergent abilities in deep learning models.

A new framework enhances graph pre-training by selecting and weighing tasks effectively.

A new method enhances accuracy in predicting protein-ligand interactions.

A new approach improves peptide drug design using AI and a comprehensive dataset.

Introducing Prompt-DDG to enhance understanding of protein mutation effects.

CBGBench offers a structured approach to enhance drug design and evaluation.

New framework improves knowledge distillation by focusing on hard samples.

FastFiD improves ODQA efficiency by selecting key sentences for quicker answers.

Exploring the efficiency and adaptability of language models through modular design.

A new method combines autoregressive and diffusion models for better media generation.