Our research explores the reliability, safety, and fairness of language and multimodal models in real-world scenarios. We aim to develop NLP systems that are not only accurate but also robust and trustworthy across diverse users, domains, and languages. 

Bias & Fairness / Robustness / Trustworthy AI / Multimodal Reliability / Safety & Alignment 

Personalization lies at the heart of modern AI applications. We focus on building recommendation systems that adapt to evolving user behavior and preferences, with applications spanning e-commerce, content platforms, and education. 

Sequential Recommendation / LLM-based Recommendation / AI for Education