Yuyang Bai  (白雨洋)

Hi there, thanks for visiting my website! I am a junior at Xi'an Jiaotong University, majoring in Artificial Intelligence. I'm looking for Ph.D. positions starting in 25Fall.

I am interested in reasoning in natural language processing and how to evaluate and improve the knowledge ability of large language models to better address real-world problems. I'm a member of LUD lab, the premiere undergraduate research group @ XJTU, advised by Prof. Minnan Luo. I have interned at UW NLP with Ph.D. student Shangbin Feng and Prof. Yulia Tsvetkov. I was also a visiting student at the University of Notre dame, working with Ph.D. student Qingkai Zeng, Zhaoxuan Tan and Prof. Meng Jiang. I am now working with Ph.D. student Bernal Jiménez Gutiérrez and Prof. Yu Su. My previous research includes social network analysis and knowledge graphs.

Email:  yuyangbai2002 [at] gmail [dot] com  /  1206944633 [at] stu [dot] xjtu [dot] edu [dot] cn

Email  /  CV  /  Google Scholar  /  Semantic Scholar  /  Twitter  /  Github

My research interests include:

• Interpreting and enhancing the knowledge & reasoning ability of Large Language Models
• NLP & social network analysis for fairness and common good

We propose KGQuiz, a knowledge-intensive benchmark to evaluate the generalizability of LLM knowledge abilities across knowledge domains and progressively complex task formats.

In this work, we introduce Chain-of-Layer (CoL), a novel framework for taxonomy induction. By leveraging the hierarchical format instruction (HF) and incorporating an Ensemble-based Ranking Filter, CoL breaks down the task into selecting relevant candidates and gradually building the taxonomy from top to bottom and significantly reduces hallucination and improves structural accuracy.

In this work, we introduce CodeTaxo, a novel approach that leverages large language models through code language prompts to capture the taxonomic structure.

We propose Knowledge Card, a community-driven initiative to empower black-box LLMs with modular and collaborative knowledge. By incorporating the outputs of independently trained, small, and specialized LMs, we make LLMs better knowledge models by empowering them with temporal knowledge update, multi-domain knowledge synthesis, and continued improvement through collective efforts.

We propose a simple, easy-to-use, shenanigan-free summarization factuality evaluation model by augmenting language models with factual knowledge from knowledge bases.

We propose MVSD, a novel Multi-View Spoiler Detection framework that takes into account the external knowledge about movies and user activities on movie review platforms.

We present Twibot-22, the largest graph-based Twibot bot detection benchmark to date, which provides diversified entities and relations in Twittersphere and has considerably better annotation quality.

Template courtesy: Jon Barron.