Yang Liu

Researcher at ByteDance Research working on trustworthiness and alignment of large language models, particularly focusing on safety evaluation and systematic frameworks for assessing LLM behavior across multiple dimensions.

Sources in this wiki

• Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models' Alignment

Topics

• LLM Safety and Adversarial Robustness, LLM Alignment, Evaluation metrics for language models